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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
microsoft/BiomedVLP-CXR-BERT-specialized | 2022-07-11T14:52:06.000Z | [
"transformers",
"pytorch",
"cxr-bert",
"feature-extraction",
"exbert",
"fill-mask",
"custom_code",
"en",
"arxiv:2204.09817",
"arxiv:2103.00020",
"arxiv:2002.05709",
"license:mit",
"region:us"
] | fill-mask | microsoft | null | null | microsoft/BiomedVLP-CXR-BERT-specialized | 14 | 7,684 | transformers | 2022-05-11T17:20:52 | ---
language: en
tags:
- exbert
license: mit
pipeline_tag: fill-mask
widget:
- text: "Left pleural effusion with adjacent [MASK]."
example_title: "Radiology 1"
- text: "Heart size normal and lungs are [MASK]."
example_title: "Radiology 2"
inference: false
---
# CXR-BERT-specialized
[CXR-BERT](https://arxiv.org/abs/2204.09817) is a chest X-ray (CXR) domain-specific language model that makes use of an improved vocabulary, novel pretraining procedure, weight regularization, and text augmentations. The resulting model demonstrates improved performance on radiology natural language inference, radiology masked language model token prediction, and downstream vision-language processing tasks such as zero-shot phrase grounding and image classification.
First, we pretrain [**CXR-BERT-general**](https://huggingface.co/microsoft/BiomedVLP-CXR-BERT-general) from a randomly initialized BERT model via Masked Language Modeling (MLM) on abstracts [PubMed](https://pubmed.ncbi.nlm.nih.gov/) and clinical notes from the publicly-available [MIMIC-III](https://physionet.org/content/mimiciii/1.4/) and [MIMIC-CXR](https://physionet.org/content/mimic-cxr/). In that regard, the general model is expected be applicable for research in clinical domains other than the chest radiology through domain specific fine-tuning.
**CXR-BERT-specialized** is continually pretrained from CXR-BERT-general to further specialize in the chest X-ray domain. At the final stage, CXR-BERT is trained in a multi-modal contrastive learning framework, similar to the [CLIP](https://arxiv.org/abs/2103.00020) framework. The latent representation of [CLS] token is utilized to align text/image embeddings.
## Model variations
| Model | Model identifier on HuggingFace | Vocabulary | Note |
| ------------------------------------------------- | ----------------------------------------------------------------------------------------------------------- | -------------- | --------------------------------------------------------- |
| CXR-BERT-general | [microsoft/BiomedVLP-CXR-BERT-general](https://huggingface.co/microsoft/BiomedVLP-CXR-BERT-general) | PubMed & MIMIC | Pretrained for biomedical literature and clinical domains |
| CXR-BERT-specialized (after multi-modal training) | [microsoft/BiomedVLP-CXR-BERT-specialized](https://huggingface.co/microsoft/BiomedVLP-CXR-BERT-specialized) | PubMed & MIMIC | Pretrained for chest X-ray domain |
## Image model
**CXR-BERT-specialized** is jointly trained with a ResNet-50 image model in a multi-modal contrastive learning framework. Prior to multi-modal learning, the image model is pre-trained on the same set of images in MIMIC-CXR using [SimCLR](https://arxiv.org/abs/2002.05709). The corresponding model definition and its loading functions can be accessed through our [HI-ML-Multimodal](https://github.com/microsoft/hi-ml/blob/main/hi-ml-multimodal/src/health_multimodal/image/model/model.py) GitHub repository. The joint image and text model, namely [BioViL](https://arxiv.org/abs/2204.09817), can be used in phrase grounding applications as shown in this python notebook [example](https://mybinder.org/v2/gh/microsoft/hi-ml/HEAD?labpath=hi-ml-multimodal%2Fnotebooks%2Fphrase_grounding.ipynb). Additionally, please check the [MS-CXR benchmark](https://physionet.org/content/ms-cxr/0.1/) for a more systematic evaluation of joint image and text models in phrase grounding tasks.
## Citation
The corresponding manuscript is accepted to be presented at the [**European Conference on Computer Vision (ECCV) 2022**](https://eccv2022.ecva.net/)
```bibtex
@misc{https://doi.org/10.48550/arxiv.2204.09817,
doi = {10.48550/ARXIV.2204.09817},
url = {https://arxiv.org/abs/2204.09817},
author = {Boecking, Benedikt and Usuyama, Naoto and Bannur, Shruthi and Castro, Daniel C. and Schwaighofer, Anton and Hyland, Stephanie and Wetscherek, Maria and Naumann, Tristan and Nori, Aditya and Alvarez-Valle, Javier and Poon, Hoifung and Oktay, Ozan},
title = {Making the Most of Text Semantics to Improve Biomedical Vision-Language Processing},
publisher = {arXiv},
year = {2022},
}
```
## Model Use
### Intended Use
This model is intended to be used solely for (I) future research on visual-language processing and (II) reproducibility of the experimental results reported in the reference paper.
#### Primary Intended Use
The primary intended use is to support AI researchers building on top of this work. CXR-BERT and its associated models should be helpful for exploring various clinical NLP & VLP research questions, especially in the radiology domain.
#### Out-of-Scope Use
**Any** deployed use case of the model --- commercial or otherwise --- is currently out of scope. Although we evaluated the models using a broad set of publicly-available research benchmarks, the models and evaluations are not intended for deployed use cases. Please refer to [the associated paper](https://arxiv.org/abs/2204.09817) for more details.
### How to use
Here is how to use this model to extract radiological sentence embeddings and obtain their cosine similarity in the joint space (image and text):
```python
import torch
from transformers import AutoModel, AutoTokenizer
# Load the model and tokenizer
url = "microsoft/BiomedVLP-CXR-BERT-specialized"
tokenizer = AutoTokenizer.from_pretrained(url, trust_remote_code=True)
model = AutoModel.from_pretrained(url, trust_remote_code=True)
# Input text prompts (e.g., reference, synonym, contradiction)
text_prompts = ["There is no pneumothorax or pleural effusion",
"No pleural effusion or pneumothorax is seen",
"The extent of the pleural effusion is constant."]
# Tokenize and compute the sentence embeddings
tokenizer_output = tokenizer.batch_encode_plus(batch_text_or_text_pairs=text_prompts,
add_special_tokens=True,
padding='longest',
return_tensors='pt')
embeddings = model.get_projected_text_embeddings(input_ids=tokenizer_output.input_ids,
attention_mask=tokenizer_output.attention_mask)
# Compute the cosine similarity of sentence embeddings obtained from input text prompts.
sim = torch.mm(embeddings, embeddings.t())
```
## Data
This model builds upon existing publicly-available datasets:
- [PubMed](https://pubmed.ncbi.nlm.nih.gov/)
- [MIMIC-III](https://physionet.org/content/mimiciii/)
- [MIMIC-CXR](https://physionet.org/content/mimic-cxr/)
These datasets reflect a broad variety of sources ranging from biomedical abstracts to intensive care unit notes to chest X-ray radiology notes. The radiology notes are accompanied with their associated chest x-ray DICOM images in MIMIC-CXR dataset.
## Performance
We demonstrate that this language model achieves state-of-the-art results in radiology natural language inference through its improved vocabulary and novel language pretraining objective leveraging semantics and discourse characteristics in radiology reports.
A highlight of comparison to other common models, including [ClinicalBERT](https://aka.ms/clinicalbert) and [PubMedBERT](https://aka.ms/pubmedbert):
| | RadNLI accuracy (MedNLI transfer) | Mask prediction accuracy | Avg. # tokens after tokenization | Vocabulary size |
| ----------------------------------------------- | :-------------------------------: | :----------------------: | :------------------------------: | :-------------: |
| RadNLI baseline | 53.30 | - | - | - |
| ClinicalBERT | 47.67 | 39.84 | 78.98 (+38.15%) | 28,996 |
| PubMedBERT | 57.71 | 35.24 | 63.55 (+11.16%) | 28,895 |
| CXR-BERT (after Phase-III) | 60.46 | 77.72 | 58.07 (+1.59%) | 30,522 |
| **CXR-BERT (after Phase-III + Joint Training)** | **65.21** | **81.58** | **58.07 (+1.59%)** | 30,522 |
CXR-BERT also contributes to better vision-language representation learning through its improved text encoding capability. Below is the zero-shot phrase grounding performance on the **MS-CXR** dataset, which evaluates the quality of image-text latent representations.
| Vision–Language Pretraining Method | Text Encoder | MS-CXR Phrase Grounding (Avg. CNR Score) |
| ---------------------------------- | ------------ | :--------------------------------------: |
| Baseline | ClinicalBERT | 0.769 |
| Baseline | PubMedBERT | 0.773 |
| ConVIRT | ClinicalBERT | 0.818 |
| GLoRIA | ClinicalBERT | 0.930 |
| **BioViL** | **CXR-BERT** | **1.027** |
| **BioViL-L** | **CXR-BERT** | **1.142** |
Additional details about performance can be found in the corresponding paper, [Making the Most of Text Semantics to Improve Biomedical Vision-Language Processing](https://arxiv.org/abs/2204.09817).
## Limitations
This model was developed using English corpora, and thus can be considered English-only.
## Further information
Please refer to the corresponding paper, ["Making the Most of Text Semantics to Improve Biomedical Vision-Language Processing", ECCV'22](https://arxiv.org/abs/2204.09817) for additional details on the model training and evaluation.
For additional inference pipelines with CXR-BERT, please refer to the [HI-ML-Multimodal GitHub](https://aka.ms/biovil-code) repository.
| 10,416 | [
[
-0.0222015380859375,
-0.053070068359375,
0.039825439453125,
0.003467559814453125,
-0.028778076171875,
-0.01092529296875,
-0.00850677490234375,
-0.033905029296875,
0.003665924072265625,
0.0264892578125,
-0.0271759033203125,
-0.05120849609375,
-0.05865478515625,
0.0055694580078125,
-0.0085601806640625,
0.06292724609375,
0.0023021697998046875,
0.022735595703125,
-0.0192718505859375,
-0.02227783203125,
-0.022613525390625,
-0.052337646484375,
-0.033355712890625,
-0.0235137939453125,
0.0278472900390625,
-0.00913238525390625,
0.06829833984375,
0.053314208984375,
0.048736572265625,
0.019073486328125,
-0.004604339599609375,
0.0032329559326171875,
-0.01464080810546875,
-0.018829345703125,
0.000006020069122314453,
-0.0174560546875,
-0.0261688232421875,
-0.00853729248046875,
0.043060302734375,
0.05023193359375,
0.007457733154296875,
-0.0219573974609375,
0.00012445449829101562,
0.037261962890625,
-0.029998779296875,
0.01372528076171875,
-0.03045654296875,
0.00836181640625,
0.0058746337890625,
-0.0214080810546875,
-0.0287628173828125,
-0.0104522705078125,
0.0362548828125,
-0.04534912109375,
0.0060272216796875,
0.0124053955078125,
0.10736083984375,
0.01541900634765625,
-0.0233306884765625,
-0.0255279541015625,
-0.00856781005859375,
0.0684814453125,
-0.05712890625,
0.044647216796875,
0.01544189453125,
0.0036640167236328125,
0.0095977783203125,
-0.07952880859375,
-0.042755126953125,
-0.02252197265625,
-0.0303955078125,
0.0218048095703125,
-0.036529541015625,
0.0233001708984375,
0.007572174072265625,
0.0224609375,
-0.056304931640625,
-0.0029697418212890625,
-0.020904541015625,
-0.0233154296875,
0.0242919921875,
-0.0004284381866455078,
0.03497314453125,
-0.036651611328125,
-0.043121337890625,
-0.0271759033203125,
-0.036376953125,
-0.004138946533203125,
-0.01690673828125,
0.012359619140625,
-0.034637451171875,
0.038665771484375,
0.0011835098266601562,
0.0635986328125,
0.00807952880859375,
-0.0229949951171875,
0.057525634765625,
-0.0298004150390625,
-0.032379150390625,
-0.00632476806640625,
0.07208251953125,
0.030975341796875,
0.00435638427734375,
-0.003849029541015625,
-0.0009307861328125,
0.013946533203125,
-0.0015611648559570312,
-0.05694580078125,
-0.017181396484375,
0.0171356201171875,
-0.0662841796875,
-0.0289764404296875,
0.0066375732421875,
-0.03289794921875,
-0.003253936767578125,
-0.0222625732421875,
0.05078125,
-0.05682373046875,
-0.006412506103515625,
0.0236358642578125,
-0.01329803466796875,
0.0224609375,
0.005748748779296875,
-0.0244598388671875,
0.021636962890625,
0.0286865234375,
0.0645751953125,
-0.008758544921875,
-0.005565643310546875,
-0.0291290283203125,
-0.00949859619140625,
-0.0014896392822265625,
0.05206298828125,
-0.039398193359375,
-0.0260162353515625,
-0.002254486083984375,
0.017120361328125,
-0.01947021484375,
-0.032958984375,
0.042449951171875,
-0.026092529296875,
0.0341796875,
-0.007137298583984375,
-0.047271728515625,
-0.01467132568359375,
0.0251007080078125,
-0.034820556640625,
0.07452392578125,
0.005939483642578125,
-0.0697021484375,
0.017425537109375,
-0.03564453125,
-0.0231475830078125,
-0.0101318359375,
-0.040374755859375,
-0.04852294921875,
-0.0037670135498046875,
0.039581298828125,
0.048553466796875,
-0.0139007568359375,
0.02294921875,
-0.00974273681640625,
-0.0087432861328125,
0.027313232421875,
-0.00787353515625,
0.061431884765625,
0.01409912109375,
-0.0302886962890625,
0.0029163360595703125,
-0.06256103515625,
0.01036834716796875,
0.005878448486328125,
-0.0086822509765625,
-0.03106689453125,
-0.005199432373046875,
0.0210113525390625,
0.0406494140625,
0.0106658935546875,
-0.045166015625,
-0.004665374755859375,
-0.0367431640625,
0.0251617431640625,
0.0350341796875,
0.0027256011962890625,
0.0181121826171875,
-0.049072265625,
0.035614013671875,
0.0160675048828125,
0.0010890960693359375,
-0.029632568359375,
-0.0362548828125,
-0.04644775390625,
-0.057525634765625,
0.019378662109375,
0.044677734375,
-0.05682373046875,
0.043426513671875,
-0.0255126953125,
-0.03533935546875,
-0.044342041015625,
-0.0205078125,
0.060211181640625,
0.0535888671875,
0.0491943359375,
-0.023040771484375,
-0.048004150390625,
-0.07391357421875,
-0.001312255859375,
0.003177642822265625,
0.0048675537109375,
0.037689208984375,
0.0270538330078125,
-0.0229339599609375,
0.06414794921875,
-0.050506591796875,
-0.04107666015625,
-0.01019287109375,
0.0206146240234375,
0.0183563232421875,
0.043365478515625,
0.04669189453125,
-0.054901123046875,
-0.043060302734375,
0.0036830902099609375,
-0.07305908203125,
-0.003826141357421875,
-0.02056884765625,
-0.00492095947265625,
0.025482177734375,
0.0579833984375,
-0.0285186767578125,
0.0374755859375,
0.05084228515625,
-0.022064208984375,
0.0288848876953125,
-0.0267791748046875,
0.0077972412109375,
-0.09674072265625,
0.0195159912109375,
0.007480621337890625,
-0.0255889892578125,
-0.035552978515625,
0.006885528564453125,
0.00434112548828125,
-0.007579803466796875,
-0.027252197265625,
0.0513916015625,
-0.056732177734375,
0.018157958984375,
-0.0009436607360839844,
0.0141448974609375,
0.00969696044921875,
0.0345458984375,
0.03369140625,
0.039581298828125,
0.05517578125,
-0.031982421875,
-0.0130615234375,
0.031768798828125,
-0.024810791015625,
0.033416748046875,
-0.07366943359375,
0.007598876953125,
-0.01242828369140625,
0.0160675048828125,
-0.07177734375,
0.00931549072265625,
0.01108551025390625,
-0.04742431640625,
0.036865234375,
-0.0015468597412109375,
-0.033111572265625,
-0.0225830078125,
-0.023284912109375,
0.02313232421875,
0.043792724609375,
-0.02679443359375,
0.041168212890625,
0.019989013671875,
-0.0025310516357421875,
-0.053466796875,
-0.061279296875,
0.005157470703125,
0.00695037841796875,
-0.061920166015625,
0.057525634765625,
-0.0099945068359375,
-0.00402069091796875,
0.017181396484375,
0.01247406005859375,
-0.0126800537109375,
-0.0165863037109375,
0.0236053466796875,
0.033050537109375,
-0.030670166015625,
0.007022857666015625,
-0.0004284381866455078,
-0.00656890869140625,
-0.01103973388671875,
-0.0103607177734375,
0.04132080078125,
-0.01129150390625,
-0.022064208984375,
-0.048126220703125,
0.0288543701171875,
0.033599853515625,
-0.00960540771484375,
0.054473876953125,
0.048004150390625,
-0.032989501953125,
0.02874755859375,
-0.0517578125,
-0.004642486572265625,
-0.03271484375,
0.047637939453125,
-0.027618408203125,
-0.0721435546875,
0.033538818359375,
0.0173797607421875,
-0.0207977294921875,
0.034027099609375,
0.049713134765625,
-0.007843017578125,
0.08099365234375,
0.0496826171875,
0.007762908935546875,
0.025726318359375,
-0.023040771484375,
0.0112152099609375,
-0.0665283203125,
-0.0190887451171875,
-0.03570556640625,
-0.0002932548522949219,
-0.048583984375,
-0.043975830078125,
0.037872314453125,
-0.0172882080078125,
0.00937652587890625,
0.0091094970703125,
-0.047760009765625,
0.00909423828125,
0.0278778076171875,
0.03167724609375,
0.0108642578125,
0.01232147216796875,
-0.03900146484375,
-0.0287933349609375,
-0.05242919921875,
-0.0294952392578125,
0.07403564453125,
0.0311431884765625,
0.051025390625,
-0.0128173828125,
0.06024169921875,
0.0030384063720703125,
0.0114593505859375,
-0.054229736328125,
0.028228759765625,
-0.02923583984375,
-0.03997802734375,
0.003261566162109375,
-0.01209259033203125,
-0.0870361328125,
0.0195770263671875,
-0.024444580078125,
-0.0360107421875,
0.0273284912109375,
0.004302978515625,
-0.03106689453125,
0.01303863525390625,
-0.037994384765625,
0.05889892578125,
-0.01055145263671875,
-0.018646240234375,
-0.00702667236328125,
-0.06890869140625,
0.010223388671875,
0.005828857421875,
0.0225677490234375,
0.01386260986328125,
-0.013763427734375,
0.04742431640625,
-0.04248046875,
0.06915283203125,
-0.0127105712890625,
0.014495849609375,
0.01406097412109375,
-0.0193023681640625,
0.017822265625,
-0.01007080078125,
0.0010347366333007812,
0.020965576171875,
0.0181121826171875,
-0.02740478515625,
-0.0273284912109375,
0.03997802734375,
-0.07769775390625,
-0.024017333984375,
-0.0556640625,
-0.0416259765625,
-0.0030803680419921875,
0.02667236328125,
0.040374755859375,
0.054473876953125,
-0.0189971923828125,
0.032684326171875,
0.059600830078125,
-0.05224609375,
0.0171356201171875,
0.02197265625,
-0.005031585693359375,
-0.0506591796875,
0.05401611328125,
0.01100921630859375,
0.0196075439453125,
0.05938720703125,
0.0173797607421875,
-0.018951416015625,
-0.04229736328125,
-0.0003924369812011719,
0.041229248046875,
-0.058441162109375,
-0.0184173583984375,
-0.08367919921875,
-0.0252532958984375,
-0.039154052734375,
-0.031982421875,
-0.005615234375,
-0.0290374755859375,
-0.030517578125,
0.00925445556640625,
0.0173187255859375,
0.0286712646484375,
-0.0128936767578125,
0.0219573974609375,
-0.06536865234375,
0.0233917236328125,
0.00276947021484375,
0.005886077880859375,
-0.0179290771484375,
-0.051177978515625,
-0.0159759521484375,
-0.0079193115234375,
-0.0219573974609375,
-0.050384521484375,
0.04925537109375,
0.027130126953125,
0.0579833984375,
0.0190887451171875,
-0.004734039306640625,
0.05731201171875,
-0.026824951171875,
0.05389404296875,
0.0291290283203125,
-0.06475830078125,
0.0401611328125,
-0.0224456787109375,
0.0298004150390625,
0.0325927734375,
0.038238525390625,
-0.037017822265625,
-0.0260009765625,
-0.06243896484375,
-0.0762939453125,
0.032623291015625,
0.0124053955078125,
0.007198333740234375,
-0.0243377685546875,
0.023529052734375,
-0.00395965576171875,
0.0016603469848632812,
-0.066650390625,
-0.0288238525390625,
-0.0223388671875,
-0.034454345703125,
0.006000518798828125,
-0.023162841796875,
-0.00539398193359375,
-0.0227813720703125,
0.0482177734375,
-0.0154266357421875,
0.052032470703125,
0.054595947265625,
-0.032379150390625,
0.01299285888671875,
0.004009246826171875,
0.052978515625,
0.0418701171875,
-0.0243682861328125,
0.00595855712890625,
0.0010614395141601562,
-0.040740966796875,
-0.01148223876953125,
0.0195159912109375,
0.00323486328125,
0.018463134765625,
0.044525146484375,
0.052001953125,
0.02691650390625,
-0.05242919921875,
0.0584716796875,
-0.021575927734375,
-0.028594970703125,
-0.021240234375,
-0.019683837890625,
0.00004476308822631836,
0.006977081298828125,
0.0258026123046875,
0.0140380859375,
0.0033969879150390625,
-0.02276611328125,
0.035888671875,
0.03289794921875,
-0.048675537109375,
-0.01324462890625,
0.042816162109375,
-0.00307464599609375,
0.01236724853515625,
0.041015625,
0.0019931793212890625,
-0.033905029296875,
0.042022705078125,
0.047149658203125,
0.058135986328125,
-0.0005183219909667969,
0.0020923614501953125,
0.0487060546875,
0.0195770263671875,
0.00771331787109375,
0.030853271484375,
0.01377105712890625,
-0.0615234375,
-0.02850341796875,
-0.030853271484375,
-0.002170562744140625,
0.0022029876708984375,
-0.062286376953125,
0.03216552734375,
-0.038360595703125,
-0.01392364501953125,
0.011383056640625,
-0.0018491744995117188,
-0.058349609375,
0.0372314453125,
0.0155181884765625,
0.07244873046875,
-0.05645751953125,
0.076904296875,
0.0599365234375,
-0.050445556640625,
-0.0618896484375,
-0.005870819091796875,
0.0001844167709350586,
-0.0794677734375,
0.0633544921875,
0.01226806640625,
0.00473785400390625,
-0.0002925395965576172,
-0.032379150390625,
-0.0706787109375,
0.08990478515625,
0.006610870361328125,
-0.03692626953125,
-0.01172637939453125,
0.001476287841796875,
0.042816162109375,
-0.037811279296875,
0.0229034423828125,
0.00982666015625,
0.0191802978515625,
0.0093536376953125,
-0.07525634765625,
0.0231475830078125,
-0.0266571044921875,
0.0137786865234375,
-0.01366424560546875,
-0.043243408203125,
0.06353759765625,
-0.017120361328125,
-0.006961822509765625,
0.0199432373046875,
0.049285888671875,
0.0272369384765625,
0.00864410400390625,
0.0267486572265625,
0.059661865234375,
0.048492431640625,
-0.0084381103515625,
0.092529296875,
-0.029266357421875,
0.026214599609375,
0.0716552734375,
-0.0014629364013671875,
0.057098388671875,
0.0362548828125,
-0.0074920654296875,
0.055450439453125,
0.0382080078125,
-0.0058441162109375,
0.05242919921875,
-0.004657745361328125,
-0.003047943115234375,
-0.0109710693359375,
-0.00878143310546875,
-0.050323486328125,
0.02923583984375,
0.02886962890625,
-0.05560302734375,
-0.0046539306640625,
0.00974273681640625,
0.034027099609375,
0.00013387203216552734,
-0.00041937828063964844,
0.056243896484375,
0.01447296142578125,
-0.03399658203125,
0.07080078125,
-0.0059814453125,
0.06610107421875,
-0.0526123046875,
0.005828857421875,
0.0021343231201171875,
0.0266571044921875,
-0.01218414306640625,
-0.043792724609375,
0.02667236328125,
-0.025238037109375,
-0.01511383056640625,
-0.00585174560546875,
0.0472412109375,
-0.0257720947265625,
-0.04150390625,
0.02960205078125,
0.031341552734375,
0.0008115768432617188,
0.01009368896484375,
-0.08416748046875,
0.0228271484375,
0.0009403228759765625,
-0.0116119384765625,
0.033111572265625,
0.03143310546875,
0.0032863616943359375,
0.04052734375,
0.0428466796875,
0.0022563934326171875,
-0.00038814544677734375,
0.005695343017578125,
0.07720947265625,
-0.0421142578125,
-0.042022705078125,
-0.055267333984375,
0.048980712890625,
-0.01146697998046875,
-0.020477294921875,
0.041595458984375,
0.0435791015625,
0.05938720703125,
-0.004779815673828125,
0.06768798828125,
-0.01546478271484375,
0.033660888671875,
-0.046142578125,
0.049468994140625,
-0.055938720703125,
0.0030345916748046875,
-0.04180908203125,
-0.034759521484375,
-0.060394287109375,
0.06890869140625,
-0.01451873779296875,
0.0069580078125,
0.077880859375,
0.082275390625,
-0.00284576416015625,
-0.033966064453125,
0.02960205078125,
0.0292510986328125,
0.0196075439453125,
0.032135009765625,
0.03094482421875,
-0.05401611328125,
0.042755126953125,
-0.028228759765625,
-0.028564453125,
-0.01336669921875,
-0.08026123046875,
-0.069091796875,
-0.054901123046875,
-0.056121826171875,
-0.047607421875,
0.01323699951171875,
0.0787353515625,
0.07080078125,
-0.0560302734375,
-0.005275726318359375,
0.00849151611328125,
-0.01288604736328125,
-0.0229034423828125,
-0.0167999267578125,
0.05499267578125,
-0.0298919677734375,
-0.03704833984375,
0.00310516357421875,
0.01580810546875,
0.00839996337890625,
0.0017185211181640625,
-0.005176544189453125,
-0.042266845703125,
0.00321197509765625,
0.048095703125,
0.01617431640625,
-0.06427001953125,
-0.0216827392578125,
0.0240936279296875,
-0.02520751953125,
0.027435302734375,
0.043182373046875,
-0.06463623046875,
0.04779052734375,
0.03662109375,
0.040679931640625,
0.032318115234375,
-0.0170135498046875,
0.02978515625,
-0.061309814453125,
0.0254974365234375,
0.01136016845703125,
0.0360107421875,
0.03515625,
-0.0301361083984375,
0.024627685546875,
0.0247802734375,
-0.041473388671875,
-0.056915283203125,
-0.00989532470703125,
-0.10107421875,
-0.0299072265625,
0.0811767578125,
-0.018157958984375,
-0.0192718505859375,
-0.007190704345703125,
-0.0274200439453125,
0.0269622802734375,
-0.0160369873046875,
0.041290283203125,
0.039398193359375,
-0.0131683349609375,
-0.033416748046875,
-0.0199127197265625,
0.0290374755859375,
0.03350830078125,
-0.05126953125,
-0.0325927734375,
0.0262451171875,
0.0374755859375,
0.019683837890625,
0.059234619140625,
-0.04058837890625,
0.0229949951171875,
-0.0169830322265625,
0.0224761962890625,
0.0004391670227050781,
0.0107421875,
-0.032470703125,
0.0056915283203125,
0.0014553070068359375,
-0.0171661376953125
]
] |
boomerchan/Magpie-13b | 2023-09-17T22:54:22.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"not-for-all-audiences",
"en",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | boomerchan | null | null | boomerchan/Magpie-13b | 7 | 7,684 | transformers | 2023-09-11T20:29:03 | ---
license: llama2
pipeline_tag: text-generation
language:
- en
tags:
- not-for-all-audiences
---
<center><h1>Magpie 13b</h1></center>
Magpie-13b is a Llama 2 merge made specifically for SillyTavern roleplays. This is the final result of approximately 2 weeks of merging and testing, and in my opinion it is the most complementary 13b merge of the included models/loras for this use-case.
This repo is for the full model weights. <a href="https://huggingface.co/boomerchan/Magpie-13b-GGUF">Click here for gguf quants</a>. Massive thanks to TheBloke for the <a href="https://huggingface.co/TheBloke/Magpie-13B-GPTQ">GPTQ version</a>.
<h2>Prompt template: Alpaca</h2>
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{prompt}
### Response:
```
<h2>My SillyTavern instruct mode prompt:</h2>
```
Write an engaging response from {{char}} to {{user}} in a descriptive writing style. Express {{char}}'s actions and sensory perceptions in vivid detail. Proactively advance the conversation with narrative prose as {{char}}. Use informal language when composing dialogue, taking into account {{char}}'s personality and communication style.
### Instruction:
{prompt}
### Response:
```
I have included 2 .json files in this repo. `Magpie.json` is an instruct mode preset which you can import into SillyTavern VIA the AI Response Formatting tab. `Magpie (Spicy).json` is a sampler preset for Oobabooga which you can import into SillyTavern VIA the AI Response Configuration tab. If you have any questions, please direct them to the SillyTavern Discord server. glhf <3
Model weights included in this merge:
- Chronos 13b
- Nous Hermes L2 13b
- OrcaPlatypus2 13b
- Spicyboros 2.2 13b
- Kimiko v2 13b Lora
- limarp L2 v2 13b Lora | 1,807 | [
[
-0.045989990234375,
-0.05230712890625,
0.0201263427734375,
0.015533447265625,
-0.0136260986328125,
-0.004146575927734375,
-0.006374359130859375,
-0.05145263671875,
0.03564453125,
0.037078857421875,
-0.055389404296875,
-0.00530242919921875,
-0.03826904296875,
0.011138916015625,
-0.01142120361328125,
0.083984375,
0.034393310546875,
-0.016845703125,
-0.0007390975952148438,
0.007205963134765625,
-0.05841064453125,
-0.03125,
-0.03240966796875,
-0.030609130859375,
0.041351318359375,
0.039794921875,
0.0743408203125,
0.0204620361328125,
0.0224761962890625,
0.0272979736328125,
-0.026031494140625,
0.061920166015625,
-0.050811767578125,
0.0022449493408203125,
-0.0276947021484375,
-0.0298004150390625,
-0.069091796875,
0.007659912109375,
0.036834716796875,
0.00579071044921875,
-0.03302001953125,
0.0171661376953125,
-0.0136260986328125,
0.025390625,
-0.040679931640625,
0.0194549560546875,
-0.01555633544921875,
0.00321197509765625,
-0.018707275390625,
0.0023555755615234375,
-0.0213775634765625,
-0.036529541015625,
-0.023651123046875,
-0.054168701171875,
-0.0123443603515625,
0.03271484375,
0.07720947265625,
0.020751953125,
-0.039459228515625,
-0.034881591796875,
-0.03961181640625,
0.040374755859375,
-0.06268310546875,
-0.00940704345703125,
0.032318115234375,
0.0367431640625,
-0.035736083984375,
-0.05133056640625,
-0.0684814453125,
-0.0115966796875,
-0.01291656494140625,
0.0116729736328125,
-0.0537109375,
-0.021453857421875,
0.01473236083984375,
0.0467529296875,
-0.04638671875,
0.003566741943359375,
-0.056182861328125,
-0.00807952880859375,
0.05218505859375,
0.031585693359375,
0.0252685546875,
-0.040252685546875,
-0.0298004150390625,
-0.0227508544921875,
-0.041839599609375,
0.0114288330078125,
0.054290771484375,
0.0218658447265625,
-0.018524169921875,
0.070068359375,
0.0019702911376953125,
0.055999755859375,
0.04803466796875,
-0.0097503662109375,
0.0212554931640625,
-0.043914794921875,
-0.0277252197265625,
-0.00860595703125,
0.0809326171875,
0.0284423828125,
0.0146942138671875,
-0.0072174072265625,
-0.0071563720703125,
-0.0158538818359375,
0.04010009765625,
-0.060791015625,
-0.00812530517578125,
0.00963592529296875,
-0.00308990478515625,
-0.0330810546875,
0.021209716796875,
-0.056915283203125,
-0.03564453125,
0.00878143310546875,
0.04107666015625,
-0.042816162109375,
-0.021697998046875,
0.0264129638671875,
0.0034923553466796875,
0.0318603515625,
0.0177459716796875,
-0.06390380859375,
0.0267791748046875,
0.04351806640625,
0.03863525390625,
0.0288543701171875,
-0.0140838623046875,
-0.05609130859375,
0.002979278564453125,
-0.0355224609375,
0.0645751953125,
-0.01629638671875,
-0.0428466796875,
-0.0231781005859375,
0.01617431640625,
-0.017333984375,
-0.0457763671875,
0.03619384765625,
-0.04046630859375,
0.0433349609375,
-0.03704833984375,
-0.0153350830078125,
-0.05169677734375,
0.0175628662109375,
-0.056854248046875,
0.06939697265625,
0.0206451416015625,
-0.040313720703125,
0.006877899169921875,
-0.045745849609375,
-0.00479888916015625,
0.01294708251953125,
0.017242431640625,
-0.0237274169921875,
0.0016956329345703125,
0.015625,
0.0271148681640625,
-0.038330078125,
0.003459930419921875,
-0.02197265625,
-0.04510498046875,
0.0168609619140625,
0.0170745849609375,
0.0648193359375,
0.0237274169921875,
-0.0169525146484375,
0.00997161865234375,
-0.03961181640625,
0.004184722900390625,
0.0316162109375,
0.0033473968505859375,
0.00923919677734375,
-0.0161895751953125,
0.0007205009460449219,
0.001499176025390625,
0.030731201171875,
-0.004238128662109375,
0.0701904296875,
-0.0179901123046875,
0.032867431640625,
0.03302001953125,
0.0027484893798828125,
0.036834716796875,
-0.046112060546875,
0.04486083984375,
-0.01490020751953125,
0.0261688232421875,
0.0250244140625,
-0.05322265625,
-0.050872802734375,
-0.0386962890625,
0.0224456787109375,
0.05609130859375,
-0.04962158203125,
0.05230712890625,
0.0012254714965820312,
-0.06591796875,
-0.02056884765625,
0.005931854248046875,
0.022216796875,
0.03472900390625,
0.032318115234375,
-0.0174713134765625,
-0.034759521484375,
-0.063232421875,
0.0198822021484375,
-0.0452880859375,
0.006160736083984375,
0.02630615234375,
0.026092529296875,
-0.043121337890625,
0.055694580078125,
-0.044464111328125,
-0.047943115234375,
-0.04644775390625,
-0.0008482933044433594,
0.03399658203125,
0.02899169921875,
0.08270263671875,
-0.047332763671875,
-0.00490570068359375,
-0.001621246337890625,
-0.04901123046875,
-0.0189666748046875,
0.0078277587890625,
-0.0201873779296875,
0.0018939971923828125,
0.0032939910888671875,
-0.08062744140625,
0.02972412109375,
0.0302581787109375,
-0.04376220703125,
0.034393310546875,
-0.00930023193359375,
0.04052734375,
-0.07611083984375,
0.0034961700439453125,
-0.03302001953125,
-0.0147705078125,
-0.044891357421875,
0.028594970703125,
0.0020580291748046875,
0.0175628662109375,
-0.038818359375,
0.04571533203125,
-0.032257080078125,
0.01082611083984375,
-0.00518035888671875,
0.0040130615234375,
0.0034694671630859375,
0.020843505859375,
-0.052459716796875,
0.07220458984375,
0.0307464599609375,
-0.02056884765625,
0.05572509765625,
0.046173095703125,
-0.0011548995971679688,
0.024688720703125,
-0.05609130859375,
0.01849365234375,
0.002849578857421875,
0.0204925537109375,
-0.090087890625,
-0.0438232421875,
0.08551025390625,
-0.032684326171875,
0.024932861328125,
-0.007476806640625,
-0.056793212890625,
-0.0458984375,
-0.0305328369140625,
0.02001953125,
0.042327880859375,
-0.024688720703125,
0.075927734375,
0.0156402587890625,
-0.0088348388671875,
-0.03814697265625,
-0.07061767578125,
-0.005886077880859375,
-0.028106689453125,
-0.04412841796875,
0.0204925537109375,
-0.021453857421875,
-0.002338409423828125,
0.0064239501953125,
0.01242828369140625,
-0.0102386474609375,
-0.0218505859375,
0.034271240234375,
0.044464111328125,
-0.01458740234375,
-0.00878143310546875,
-0.0008521080017089844,
0.015655517578125,
-0.01421356201171875,
0.0231170654296875,
0.05145263671875,
-0.04742431640625,
-0.01198577880859375,
-0.02545166015625,
0.01715087890625,
0.047882080078125,
-0.01270294189453125,
0.073486328125,
0.054656982421875,
-0.0150909423828125,
0.0257720947265625,
-0.06072998046875,
0.0161590576171875,
-0.034881591796875,
0.00113677978515625,
-0.01023101806640625,
-0.044952392578125,
0.053436279296875,
0.040740966796875,
-0.0026683807373046875,
0.050079345703125,
0.046600341796875,
0.0105438232421875,
0.0477294921875,
0.0535888671875,
-0.0043792724609375,
0.0246124267578125,
-0.033477783203125,
0.028961181640625,
-0.053558349609375,
-0.0548095703125,
-0.02105712890625,
-0.0253448486328125,
-0.029144287109375,
-0.0391845703125,
0.00969696044921875,
0.03961181640625,
-0.017974853515625,
0.053314208984375,
-0.03607177734375,
0.0291595458984375,
0.024810791015625,
0.009735107421875,
0.0214080810546875,
-0.01458740234375,
0.0204620361328125,
0.004756927490234375,
-0.062225341796875,
-0.035186767578125,
0.057952880859375,
-0.005916595458984375,
0.05950927734375,
0.0469970703125,
0.055328369140625,
0.00423431396484375,
0.0211639404296875,
-0.06890869140625,
0.055877685546875,
0.00788116455078125,
-0.043487548828125,
-0.006343841552734375,
-0.02386474609375,
-0.053070068359375,
0.0165252685546875,
-0.03790283203125,
-0.06121826171875,
0.01049041748046875,
0.0191650390625,
-0.0582275390625,
0.01268768310546875,
-0.06976318359375,
0.048431396484375,
0.0019216537475585938,
-0.007030487060546875,
-0.01122283935546875,
-0.035430908203125,
0.06768798828125,
-0.00957489013671875,
0.003387451171875,
0.0019550323486328125,
-0.01201629638671875,
0.03338623046875,
-0.0416259765625,
0.047943115234375,
0.0089263916015625,
0.00043773651123046875,
0.04437255859375,
0.005397796630859375,
0.03240966796875,
0.0224761962890625,
0.0031757354736328125,
0.0157012939453125,
0.0015096664428710938,
-0.008758544921875,
-0.03662109375,
0.050384521484375,
-0.069091796875,
-0.0599365234375,
-0.05035400390625,
-0.04705810546875,
0.00702667236328125,
-0.0033092498779296875,
0.0309295654296875,
0.0167236328125,
0.006134033203125,
0.003467559814453125,
0.036407470703125,
-0.0019855499267578125,
0.0263214111328125,
0.0300750732421875,
-0.041839599609375,
-0.038543701171875,
0.03533935546875,
0.01279449462890625,
-0.00035572052001953125,
0.0014514923095703125,
0.01416778564453125,
-0.0278778076171875,
0.0004978179931640625,
-0.04107666015625,
0.034423828125,
-0.034210205078125,
-0.048736572265625,
-0.042266845703125,
-0.030731201171875,
-0.032867431640625,
0.0004165172576904297,
-0.01171875,
-0.040740966796875,
-0.03997802734375,
0.01480865478515625,
0.061279296875,
0.06915283203125,
-0.035308837890625,
0.051177978515625,
-0.0452880859375,
0.033203125,
0.041900634765625,
-0.02508544921875,
0.005474090576171875,
-0.06829833984375,
0.01357269287109375,
-0.0059661865234375,
-0.036834716796875,
-0.0882568359375,
0.0288238525390625,
0.021514892578125,
0.03643798828125,
0.0325927734375,
0.005413055419921875,
0.073974609375,
-0.03387451171875,
0.0711669921875,
0.019622802734375,
-0.0673828125,
0.0352783203125,
-0.0215606689453125,
0.01043701171875,
0.01207733154296875,
0.036895751953125,
-0.0166778564453125,
-0.0213775634765625,
-0.048126220703125,
-0.07061767578125,
0.054931640625,
0.015777587890625,
0.0022144317626953125,
-0.0110015869140625,
0.0161895751953125,
-0.0038661956787109375,
0.01120758056640625,
-0.043609619140625,
-0.019622802734375,
-0.0234527587890625,
0.00670623779296875,
0.0011930465698242188,
-0.0241851806640625,
-0.032257080078125,
-0.007312774658203125,
0.0430908203125,
0.0064849853515625,
0.0217132568359375,
-0.00984954833984375,
-0.005123138427734375,
-0.00571441650390625,
0.0167999267578125,
0.056365966796875,
0.04034423828125,
-0.0259857177734375,
-0.01348876953125,
0.0194854736328125,
-0.051605224609375,
0.006481170654296875,
0.0221099853515625,
0.01187896728515625,
-0.030364990234375,
0.033935546875,
0.028564453125,
0.0216217041015625,
-0.03546142578125,
0.049560546875,
-0.01020050048828125,
-0.020782470703125,
-0.0014629364013671875,
0.037689208984375,
0.01056671142578125,
0.057098388671875,
0.032379150390625,
0.00640869140625,
0.011688232421875,
-0.038116455078125,
-0.00937652587890625,
0.0293731689453125,
0.01035308837890625,
-0.016845703125,
0.04986572265625,
-0.00882720947265625,
-0.0160675048828125,
0.038055419921875,
-0.039215087890625,
-0.0298614501953125,
0.04058837890625,
0.042449951171875,
0.0447998046875,
-0.0093994140625,
0.007045745849609375,
0.0283203125,
0.02587890625,
-0.0118865966796875,
0.0300750732421875,
0.0023555755615234375,
-0.0213470458984375,
-0.038909912109375,
-0.0244293212890625,
-0.038330078125,
0.01302337646484375,
-0.0633544921875,
0.027435302734375,
-0.037506103515625,
-0.0277862548828125,
-0.0145721435546875,
0.0200653076171875,
-0.0426025390625,
-0.0033931732177734375,
0.004756927490234375,
0.048919677734375,
-0.0509033203125,
0.043975830078125,
0.061431884765625,
-0.04547119140625,
-0.08941650390625,
-0.01300811767578125,
0.01727294921875,
-0.0679931640625,
0.04547119140625,
-0.001819610595703125,
-0.0140380859375,
-0.0175018310546875,
-0.056488037109375,
-0.06890869140625,
0.09130859375,
0.01873779296875,
-0.033935546875,
-0.0097808837890625,
-0.0200042724609375,
0.0197296142578125,
-0.0192413330078125,
0.025909423828125,
0.0255889892578125,
0.0445556640625,
0.03314208984375,
-0.071533203125,
0.0164642333984375,
-0.0178985595703125,
-0.007293701171875,
0.0030078887939453125,
-0.0802001953125,
0.07354736328125,
-0.0288543701171875,
-0.0159759521484375,
0.04901123046875,
0.06353759765625,
0.035369873046875,
-0.0080718994140625,
0.0390625,
0.04315185546875,
0.06121826171875,
-0.0197296142578125,
0.05706787109375,
-0.01383209228515625,
0.037689208984375,
0.068359375,
-0.0166473388671875,
0.05267333984375,
0.03033447265625,
0.004467010498046875,
0.0284271240234375,
0.0411376953125,
-0.0163421630859375,
0.04815673828125,
-0.0007157325744628906,
0.0036220550537109375,
-0.0166168212890625,
0.0009207725524902344,
-0.06475830078125,
0.039642333984375,
0.0123748779296875,
0.0224456787109375,
-0.01641845703125,
-0.00927734375,
0.0146636962890625,
-0.01155853271484375,
-0.0253753662109375,
0.026763916015625,
-0.01044464111328125,
-0.032806396484375,
0.05145263671875,
-0.015045166015625,
0.061431884765625,
-0.07275390625,
-0.0038242340087890625,
-0.03680419921875,
0.004730224609375,
-0.025146484375,
-0.06048583984375,
0.0288848876953125,
-0.0009655952453613281,
-0.034332275390625,
0.004863739013671875,
0.038330078125,
-0.0288238525390625,
-0.0260467529296875,
0.00656890869140625,
0.02655029296875,
0.0128936767578125,
0.019775390625,
-0.03607177734375,
0.0311737060546875,
-0.010162353515625,
-0.0188446044921875,
0.016387939453125,
0.03662109375,
0.00312042236328125,
0.0552978515625,
0.031280517578125,
0.00395965576171875,
-0.01202392578125,
-0.0155029296875,
0.09027099609375,
-0.04339599609375,
-0.02880859375,
-0.044647216796875,
0.033599853515625,
0.0108489990234375,
-0.032928466796875,
0.045562744140625,
0.026092529296875,
0.037872314453125,
-0.0202484130859375,
0.012725830078125,
-0.036865234375,
0.01363372802734375,
-0.051361083984375,
0.0775146484375,
-0.0562744140625,
0.006732940673828125,
-0.00836944580078125,
-0.07244873046875,
0.0008373260498046875,
0.05438232421875,
-0.017486572265625,
-0.0092010498046875,
0.035858154296875,
0.07257080078125,
-0.01474761962890625,
-0.0107421875,
0.0182647705078125,
0.01444244384765625,
0.01491546630859375,
0.0709228515625,
0.073974609375,
-0.05950927734375,
0.05511474609375,
0.004276275634765625,
-0.02337646484375,
-0.03631591796875,
-0.0765380859375,
-0.0792236328125,
-0.0106353759765625,
-0.009429931640625,
-0.027618408203125,
0.008544921875,
0.0809326171875,
0.05078125,
-0.036224365234375,
-0.008636474609375,
0.038909912109375,
-0.00400543212890625,
-0.0084991455078125,
-0.0175933837890625,
0.0008563995361328125,
0.00489044189453125,
-0.054290771484375,
0.034393310546875,
0.01273345947265625,
0.0282135009765625,
0.005855560302734375,
-0.0047149658203125,
-0.0166473388671875,
0.0175628662109375,
0.024505615234375,
0.04522705078125,
-0.0726318359375,
-0.019134521484375,
0.006809234619140625,
-0.0192718505859375,
-0.012451171875,
0.02056884765625,
-0.062042236328125,
0.00025177001953125,
0.02056884765625,
0.0191650390625,
0.031707763671875,
0.0026264190673828125,
0.026458740234375,
-0.0277862548828125,
0.036407470703125,
0.0001010894775390625,
0.0230865478515625,
0.027191162109375,
-0.04290771484375,
0.049835205078125,
0.0015592575073242188,
-0.04638671875,
-0.06072998046875,
-0.002971649169921875,
-0.10888671875,
-0.0018320083618164062,
0.0765380859375,
-0.0019855499267578125,
-0.0121612548828125,
0.036376953125,
-0.0478515625,
0.01514434814453125,
-0.0294189453125,
0.043426513671875,
0.0516357421875,
-0.02081298828125,
-0.004302978515625,
-0.0296478271484375,
0.01557159423828125,
0.011932373046875,
-0.0570068359375,
-0.024261474609375,
0.040252685546875,
0.01366424560546875,
0.0294952392578125,
0.028533935546875,
-0.0025043487548828125,
0.036163330078125,
-0.01092529296875,
0.006744384765625,
-0.0455322265625,
-0.0241546630859375,
-0.00974273681640625,
0.005596160888671875,
-0.0191497802734375,
-0.00792694091796875
]
] |
hfl/chinese-alpaca-2-7b | 2023-08-25T01:06:56.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | hfl | null | null | hfl/chinese-alpaca-2-7b | 112 | 7,677 | transformers | 2023-07-31T03:53:55 | ---
license: apache-2.0
---
# Chinese-Alpaca-2-7B
**This is the full Chinese-Alpaca-2-7B model，which can be loaded directly for inference and full-parameter training.**
**Related models👇**
* Long context base models
* [Chinese-LLaMA-2-7B-16K (full model)](https://huggingface.co/ziqingyang/chinese-llama-2-7b-16k)
* [Chinese-LLaMA-2-LoRA-7B-16K (LoRA model)](https://huggingface.co/ziqingyang/chinese-llama-2-lora-7b-16k)
* [Chinese-LLaMA-2-13B-16K (full model)](https://huggingface.co/ziqingyang/chinese-llama-2-13b-16k)
* [Chinese-LLaMA-2-LoRA-13B-16K (LoRA model)](https://huggingface.co/ziqingyang/chinese-llama-2-lora-13b-16k)
* Base models
* [Chinese-LLaMA-2-7B (full model)](https://huggingface.co/ziqingyang/chinese-llama-2-7b)
* [Chinese-LLaMA-2-LoRA-7B (LoRA model)](https://huggingface.co/ziqingyang/chinese-llama-2-lora-7b)
* [Chinese-LLaMA-2-13B (full model)](https://huggingface.co/ziqingyang/chinese-llama-2-13b)
* [Chinese-LLaMA-2-LoRA-13B (LoRA model)](https://huggingface.co/ziqingyang/chinese-llama-2-lora-13b)
* Instruction/Chat models
* [Chinese-Alpaca-2-7B (full model)](https://huggingface.co/ziqingyang/chinese-alpaca-2-7b)
* [Chinese-Alpaca-2-LoRA-7B (LoRA model)](https://huggingface.co/ziqingyang/chinese-alpaca-2-lora-7b)
* [Chinese-Alpaca-2-13B (full model)](https://huggingface.co/ziqingyang/chinese-alpaca-2-13b)
* [Chinese-Alpaca-2-LoRA-13B (LoRA model)](https://huggingface.co/ziqingyang/chinese-alpaca-2-lora-13b)
# Description of Chinese-LLaMA-Alpaca-2
This project is based on the Llama-2, released by Meta, and it is the second generation of the Chinese LLaMA & Alpaca LLM project. We open-source Chinese LLaMA-2 (foundation model) and Alpaca-2 (instruction-following model). These models have been expanded and optimized with Chinese vocabulary beyond the original Llama-2. We used large-scale Chinese data for incremental pre-training, which further improved the fundamental semantic understanding of the Chinese language, resulting in a significant performance improvement compared to the first-generation models. The relevant models support a 4K context and can be expanded up to 18K+ using the NTK method.
The main contents of this project include:
* 🚀 New extended Chinese vocabulary beyond Llama-2, open-sourcing the Chinese LLaMA-2 and Alpaca-2 LLMs.
* 🚀 Open-sourced the pre-training and instruction finetuning (SFT) scripts for further tuning on user's data
* 🚀 Quickly deploy and experience the quantized LLMs on CPU/GPU of personal PC
* 🚀 Support for LLaMA ecosystems like 🤗transformers, llama.cpp, text-generation-webui, LangChain, vLLM etc.
Please refer to [https://github.com/ymcui/Chinese-LLaMA-Alpaca-2/](https://github.com/ymcui/Chinese-LLaMA-Alpaca-2/) for details. | 2,754 | [
[
-0.0298004150390625,
-0.046112060546875,
0.0202789306640625,
0.054656982421875,
-0.047515869140625,
-0.0182037353515625,
0.003368377685546875,
-0.0677490234375,
0.0269927978515625,
0.033905029296875,
-0.0418701171875,
-0.03875732421875,
-0.040435791015625,
0.0019216537475585938,
-0.0199737548828125,
0.05078125,
-0.00484466552734375,
0.01134490966796875,
0.026031494140625,
-0.027435302734375,
-0.0271759033203125,
-0.0259552001953125,
-0.054046630859375,
-0.03729248046875,
0.052825927734375,
0.005584716796875,
0.060394287109375,
0.06500244140625,
0.0297393798828125,
0.015411376953125,
-0.0186004638671875,
0.02191162109375,
-0.0187835693359375,
-0.03369140625,
0.01120758056640625,
-0.0291748046875,
-0.060546875,
-0.0011234283447265625,
0.0247802734375,
0.034149169921875,
-0.017822265625,
0.0287628173828125,
-0.0018749237060546875,
0.028533935546875,
-0.0321044921875,
0.02117919921875,
-0.03485107421875,
0.00452423095703125,
-0.0264739990234375,
0.0030364990234375,
-0.0210113525390625,
-0.0157012939453125,
-0.0089874267578125,
-0.06939697265625,
0.00036525726318359375,
-0.0071868896484375,
0.09619140625,
0.01873779296875,
-0.0413818359375,
-0.0254669189453125,
-0.0184326171875,
0.0545654296875,
-0.07012939453125,
0.01702880859375,
0.037933349609375,
0.01024627685546875,
-0.0260467529296875,
-0.055908203125,
-0.046844482421875,
-0.02349853515625,
-0.019439697265625,
0.01200103759765625,
0.0089111328125,
-0.01256561279296875,
0.007167816162109375,
0.02203369140625,
-0.033905029296875,
0.03570556640625,
-0.040130615234375,
-0.00615692138671875,
0.053619384765625,
-0.01033782958984375,
0.022430419921875,
-0.0027904510498046875,
-0.038116455078125,
-0.0150299072265625,
-0.0693359375,
0.00991058349609375,
0.0202484130859375,
0.02850341796875,
-0.047271728515625,
0.03668212890625,
-0.024444580078125,
0.04833984375,
-0.00167083740234375,
-0.032562255859375,
0.0438232421875,
-0.0301666259765625,
-0.01479339599609375,
-0.0163726806640625,
0.060028076171875,
0.023529052734375,
-0.007442474365234375,
0.0135345458984375,
-0.01441192626953125,
-0.01262664794921875,
-0.033416748046875,
-0.061126708984375,
0.0005855560302734375,
0.004543304443359375,
-0.049407958984375,
-0.0238189697265625,
0.008148193359375,
-0.03216552734375,
-0.01242828369140625,
-0.01386260986328125,
0.0251922607421875,
-0.013336181640625,
-0.0308990478515625,
0.0187835693359375,
0.007167816162109375,
0.0670166015625,
0.0267791748046875,
-0.0587158203125,
0.01030731201171875,
0.03961181640625,
0.059600830078125,
0.00513458251953125,
-0.0292205810546875,
-0.000030040740966796875,
0.0209197998046875,
-0.038665771484375,
0.052978515625,
-0.013641357421875,
-0.03289794921875,
-0.0137176513671875,
0.032012939453125,
0.01030731201171875,
-0.031646728515625,
0.0469970703125,
-0.03082275390625,
0.005542755126953125,
-0.0445556640625,
-0.0093994140625,
-0.039215087890625,
0.0199432373046875,
-0.0672607421875,
0.08636474609375,
0.005096435546875,
-0.04559326171875,
0.0171356201171875,
-0.050048828125,
-0.01500701904296875,
-0.01210784912109375,
-0.0012559890747070312,
-0.022369384765625,
-0.023529052734375,
0.0164947509765625,
0.027740478515625,
-0.046844482421875,
-0.0005617141723632812,
-0.020355224609375,
-0.03961181640625,
-0.0070037841796875,
-0.004367828369140625,
0.08538818359375,
0.0122222900390625,
-0.02142333984375,
-0.006839752197265625,
-0.065185546875,
-0.01276397705078125,
0.0567626953125,
-0.0294647216796875,
-0.0020923614501953125,
-0.0013599395751953125,
-0.00937652587890625,
0.01279449462890625,
0.050079345703125,
-0.0297393798828125,
0.0202178955078125,
-0.025115966796875,
0.030975341796875,
0.05224609375,
-0.01107025146484375,
0.007266998291015625,
-0.0288543701171875,
0.018585205078125,
0.01190948486328125,
0.0264129638671875,
-0.0037899017333984375,
-0.054351806640625,
-0.08624267578125,
-0.01708984375,
0.007587432861328125,
0.049835205078125,
-0.04833984375,
0.041748046875,
0.008056640625,
-0.056610107421875,
-0.0235595703125,
0.01006317138671875,
0.03515625,
0.020233154296875,
0.022186279296875,
-0.0216064453125,
-0.0438232421875,
-0.07659912109375,
0.0205078125,
-0.032684326171875,
-0.00717926025390625,
0.00811004638671875,
0.039764404296875,
-0.0225982666015625,
0.036529541015625,
-0.0261993408203125,
-0.0108642578125,
-0.01513671875,
-0.0118865966796875,
0.032440185546875,
0.033721923828125,
0.07269287109375,
-0.04351806640625,
-0.01284027099609375,
0.0086822509765625,
-0.052764892578125,
-0.0006961822509765625,
0.00017583370208740234,
-0.033050537109375,
0.01337432861328125,
0.002346038818359375,
-0.056182861328125,
0.030487060546875,
0.046173095703125,
-0.0165252685546875,
0.02703857421875,
-0.0068206787109375,
-0.0155181884765625,
-0.08270263671875,
0.0067291259765625,
-0.00579833984375,
0.014892578125,
-0.03173828125,
0.032379150390625,
0.01085662841796875,
0.0303955078125,
-0.05181884765625,
0.058990478515625,
-0.04339599609375,
-0.0157928466796875,
-0.0125885009765625,
0.006175994873046875,
0.019561767578125,
0.0589599609375,
0.00685882568359375,
0.0445556640625,
0.0266571044921875,
-0.040130615234375,
0.042694091796875,
0.034149169921875,
-0.02386474609375,
-0.00530242919921875,
-0.0645751953125,
0.0271453857421875,
0.00273895263671875,
0.05157470703125,
-0.0567626953125,
-0.02154541015625,
0.0469970703125,
-0.0236053466796875,
-0.001689910888671875,
0.01187896728515625,
-0.043792724609375,
-0.03997802734375,
-0.046417236328125,
0.035675048828125,
0.041229248046875,
-0.07232666015625,
0.0239410400390625,
0.00021457672119140625,
0.0160675048828125,
-0.061370849609375,
-0.07220458984375,
-0.0089111328125,
-0.0167999267578125,
-0.036376953125,
0.024871826171875,
-0.01067352294921875,
-0.0019683837890625,
-0.01386260986328125,
0.0066375732421875,
-0.007732391357421875,
0.006931304931640625,
0.0113372802734375,
0.053436279296875,
-0.0273590087890625,
-0.00994110107421875,
0.0022106170654296875,
0.00995635986328125,
-0.0081939697265625,
0.01030731201171875,
0.05047607421875,
-0.005359649658203125,
-0.0232696533203125,
-0.036865234375,
0.0092926025390625,
0.00745391845703125,
-0.0201263427734375,
0.062225341796875,
0.054718017578125,
-0.037017822265625,
0.0005025863647460938,
-0.0413818359375,
0.00630950927734375,
-0.035797119140625,
0.0260009765625,
-0.03839111328125,
-0.043365478515625,
0.0477294921875,
0.0198211669921875,
0.025360107421875,
0.04901123046875,
0.050201416015625,
0.0233612060546875,
0.07672119140625,
0.04840087890625,
-0.017578125,
0.033660888671875,
-0.027435302734375,
-0.001972198486328125,
-0.0579833984375,
-0.040191650390625,
-0.033599853515625,
-0.017242431640625,
-0.037109375,
-0.0440673828125,
-0.0006818771362304688,
0.0261688232421875,
-0.053436279296875,
0.0379638671875,
-0.046844482421875,
0.028778076171875,
0.042572021484375,
0.0238037109375,
0.0242156982421875,
0.006076812744140625,
0.00769805908203125,
0.028656005859375,
-0.02313232421875,
-0.041778564453125,
0.0770263671875,
0.029449462890625,
0.037139892578125,
0.01183319091796875,
0.03253173828125,
-0.0019359588623046875,
0.021759033203125,
-0.060211181640625,
0.0467529296875,
-0.0147552490234375,
-0.032562255859375,
-0.0063934326171875,
-0.00713348388671875,
-0.06732177734375,
0.0301055908203125,
0.01004791259765625,
-0.044281005859375,
0.003917694091796875,
-0.0036830902099609375,
-0.02484130859375,
0.01554107666015625,
-0.0276031494140625,
0.035858154296875,
-0.0310821533203125,
0.0025882720947265625,
-0.007221221923828125,
-0.047515869140625,
0.063720703125,
-0.0186309814453125,
0.006256103515625,
-0.03350830078125,
-0.0300445556640625,
0.057769775390625,
-0.041107177734375,
0.06732177734375,
-0.016998291015625,
-0.032318115234375,
0.049163818359375,
-0.0211944580078125,
0.051788330078125,
-0.0021419525146484375,
-0.021575927734375,
0.0406494140625,
-0.001224517822265625,
-0.040008544921875,
-0.0157318115234375,
0.036529541015625,
-0.08837890625,
-0.04486083984375,
-0.022003173828125,
-0.018310546875,
-0.0018072128295898438,
0.007152557373046875,
0.03692626953125,
-0.0048980712890625,
-0.004436492919921875,
0.008941650390625,
0.01407623291015625,
-0.0283660888671875,
0.03839111328125,
0.04217529296875,
-0.01318359375,
-0.0283966064453125,
0.045745849609375,
0.006885528564453125,
0.01433563232421875,
0.0251922607421875,
0.0122528076171875,
-0.01387786865234375,
-0.0343017578125,
-0.052093505859375,
0.041961669921875,
-0.0538330078125,
-0.0158538818359375,
-0.035400390625,
-0.042694091796875,
-0.027130126953125,
0.0031299591064453125,
-0.019439697265625,
-0.0362548828125,
-0.04473876953125,
-0.0183868408203125,
0.037445068359375,
0.040802001953125,
-0.005168914794921875,
0.052642822265625,
-0.04693603515625,
0.031494140625,
0.0293426513671875,
0.006977081298828125,
0.01255035400390625,
-0.061126708984375,
-0.012115478515625,
0.0217437744140625,
-0.034942626953125,
-0.058380126953125,
0.037841796875,
0.0207061767578125,
0.039520263671875,
0.044677734375,
-0.021514892578125,
0.06732177734375,
-0.020965576171875,
0.07916259765625,
0.026947021484375,
-0.052734375,
0.04241943359375,
-0.0206146240234375,
-0.0061187744140625,
0.0119476318359375,
0.00948333740234375,
-0.0275421142578125,
-0.0002684593200683594,
-0.021514892578125,
-0.056549072265625,
0.066650390625,
0.00537872314453125,
0.0156402587890625,
0.003498077392578125,
0.039459228515625,
0.01959228515625,
-0.0039215087890625,
-0.08245849609375,
-0.027618408203125,
-0.033721923828125,
-0.00502777099609375,
0.004688262939453125,
-0.02691650390625,
-0.01104736328125,
-0.0260009765625,
0.07196044921875,
-0.0164794921875,
0.01235198974609375,
-0.0004134178161621094,
0.0092620849609375,
-0.0087432861328125,
-0.021728515625,
0.048614501953125,
0.03253173828125,
-0.006816864013671875,
-0.0290679931640625,
0.03509521484375,
-0.03985595703125,
0.00881195068359375,
0.0034236907958984375,
-0.0193328857421875,
0.0006704330444335938,
0.0400390625,
0.07232666015625,
-0.0019397735595703125,
-0.046844482421875,
0.036895751953125,
0.0047454833984375,
-0.00785064697265625,
-0.044281005859375,
0.0017290115356445312,
0.02313232421875,
0.0279693603515625,
0.0179290771484375,
-0.027435302734375,
0.004405975341796875,
-0.038421630859375,
-0.0182342529296875,
0.018829345703125,
0.017242431640625,
-0.033721923828125,
0.050567626953125,
0.00867462158203125,
-0.0072784423828125,
0.033447265625,
-0.0258636474609375,
-0.0168609619140625,
0.088134765625,
0.0496826171875,
0.036529541015625,
-0.0302581787109375,
0.0086517333984375,
0.049560546875,
0.023284912109375,
-0.035858154296875,
0.028564453125,
0.006694793701171875,
-0.057769775390625,
-0.0137176513671875,
-0.051544189453125,
-0.02752685546875,
0.03094482421875,
-0.046417236328125,
0.048980712890625,
-0.04052734375,
-0.01043701171875,
-0.0179595947265625,
0.0276031494140625,
-0.042022705078125,
0.0189971923828125,
0.03302001953125,
0.07794189453125,
-0.044219970703125,
0.0865478515625,
0.044830322265625,
-0.0296173095703125,
-0.08319091796875,
-0.0296478271484375,
-0.0017719268798828125,
-0.112060546875,
0.045379638671875,
0.0194549560546875,
0.0009098052978515625,
-0.029632568359375,
-0.060546875,
-0.09368896484375,
0.12335205078125,
0.02630615234375,
-0.03704833984375,
-0.01210784912109375,
0.0144195556640625,
0.0281219482421875,
-0.018157958984375,
0.0237884521484375,
0.0526123046875,
0.035369873046875,
0.03814697265625,
-0.06988525390625,
0.01325225830078125,
-0.027862548828125,
0.0119476318359375,
-0.00881195068359375,
-0.10943603515625,
0.09783935546875,
-0.018096923828125,
-0.006683349609375,
0.052764892578125,
0.06744384765625,
0.0628662109375,
0.0119781494140625,
0.045562744140625,
0.033599853515625,
0.04534912109375,
0.005550384521484375,
0.046722412109375,
-0.0179290771484375,
0.0191192626953125,
0.0693359375,
-0.02081298828125,
0.06353759765625,
0.015655517578125,
-0.0256500244140625,
0.041839599609375,
0.08953857421875,
-0.01727294921875,
0.0274810791015625,
0.006256103515625,
-0.01494598388671875,
-0.001262664794921875,
-0.0193023681640625,
-0.060272216796875,
0.041748046875,
0.03436279296875,
-0.0243682861328125,
-0.0018415451049804688,
-0.03265380859375,
0.026885986328125,
-0.0391845703125,
-0.0208282470703125,
0.0288848876953125,
0.018463134765625,
-0.0302581787109375,
0.06036376953125,
0.0238037109375,
0.07012939453125,
-0.0556640625,
-0.00395965576171875,
-0.0386962890625,
-0.00386810302734375,
-0.022247314453125,
-0.032684326171875,
-0.004184722900390625,
0.0029087066650390625,
0.007598876953125,
0.0232391357421875,
0.04486083984375,
-0.016632080078125,
-0.057769775390625,
0.049346923828125,
0.0300445556640625,
0.0224609375,
0.0147857666015625,
-0.060272216796875,
0.013153076171875,
0.004711151123046875,
-0.058135986328125,
0.0212249755859375,
0.02178955078125,
-0.01454925537109375,
0.052520751953125,
0.055328369140625,
0.006000518798828125,
0.01507568359375,
0.00621795654296875,
0.061981201171875,
-0.046966552734375,
-0.0225830078125,
-0.057281494140625,
0.01262664794921875,
0.00033283233642578125,
-0.0251007080078125,
0.03131103515625,
0.030792236328125,
0.063720703125,
0.0034027099609375,
0.045867919921875,
-0.007160186767578125,
0.03106689453125,
-0.03289794921875,
0.03985595703125,
-0.059814453125,
0.01404571533203125,
0.0024890899658203125,
-0.0657958984375,
-0.0124359130859375,
0.051116943359375,
0.0099334716796875,
0.00963592529296875,
0.022430419921875,
0.05364990234375,
0.00937652587890625,
-0.0007739067077636719,
0.00653076171875,
0.0182037353515625,
0.03070068359375,
0.06829833984375,
0.05841064453125,
-0.05078125,
0.04449462890625,
-0.0469970703125,
-0.0179901123046875,
-0.01019287109375,
-0.06109619140625,
-0.0584716796875,
-0.0186004638671875,
-0.00386810302734375,
-0.0097808837890625,
-0.0147552490234375,
0.0675048828125,
0.056060791015625,
-0.053497314453125,
-0.03192138671875,
0.0212249755859375,
-0.0048675537109375,
-0.0079193115234375,
-0.00881195068359375,
0.044281005859375,
0.00799560546875,
-0.06396484375,
0.0254974365234375,
-0.0021343231201171875,
0.0312042236328125,
-0.01270294189453125,
-0.01910400390625,
-0.009857177734375,
0.0139923095703125,
0.05523681640625,
0.0321044921875,
-0.0728759765625,
-0.02215576171875,
-0.005344390869140625,
-0.0198822021484375,
0.0104827880859375,
0.0102081298828125,
-0.04547119140625,
-0.02703857421875,
0.020751953125,
0.0212860107421875,
0.034027099609375,
0.00004547834396362305,
0.004425048828125,
-0.0252685546875,
0.056427001953125,
-0.01140594482421875,
0.045166015625,
0.021881103515625,
-0.0191650390625,
0.07177734375,
0.01482391357421875,
-0.02337646484375,
-0.057525634765625,
0.0073089599609375,
-0.10125732421875,
-0.021453857421875,
0.08441162109375,
-0.0206146240234375,
-0.0294342041015625,
0.0301971435546875,
-0.026580810546875,
0.038330078125,
-0.01367950439453125,
0.04437255859375,
0.039337158203125,
0.0005679130554199219,
-0.00591278076171875,
-0.0438232421875,
0.005748748779296875,
0.02496337890625,
-0.06500244140625,
-0.02197265625,
0.01318359375,
0.020111083984375,
0.01084136962890625,
0.042510986328125,
-0.00923919677734375,
0.014556884765625,
-0.00588226318359375,
0.0208282470703125,
-0.00994110107421875,
0.0033130645751953125,
0.0038623809814453125,
-0.019073486328125,
0.00522613525390625,
-0.033782958984375
]
] |
naclbit/trinart_stable_diffusion_v2 | 2023-05-07T17:12:04.000Z | [
"diffusers",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | naclbit | null | null | naclbit/trinart_stable_diffusion_v2 | 314 | 7,675 | diffusers | 2022-09-08T10:18:16 | ---
inference: true
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
license: creativeml-openrail-m
---
## Please Note!
This model is NOT the 19.2M images Characters Model on TrinArt, but an improved version of the original Trin-sama Twitter bot model. This model is intended to retain the original SD's aesthetics as much as possible while nudging the model to anime/manga style.
Other TrinArt models can be found at:
https://huggingface.co/naclbit/trinart_derrida_characters_v2_stable_diffusion
https://huggingface.co/naclbit/trinart_characters_19.2m_stable_diffusion_v1
## Diffusers
The model has been ported to `diffusers` by [ayan4m1](https://huggingface.co/ayan4m1)
and can easily be run from one of the branches:
- `revision="diffusers-60k"` for the checkpoint trained on 60,000 steps,
- `revision="diffusers-95k"` for the checkpoint trained on 95,000 steps,
- `revision="diffusers-115k"` for the checkpoint trained on 115,000 steps.
For more information, please have a look at [the "Three flavors" section](#three-flavors).
## Gradio
We also support a [Gradio](https://github.com/gradio-app/gradio) web ui with diffusers to run inside a colab notebook: [](https://colab.research.google.com/drive/1RWvik_C7nViiR9bNsu3fvMR3STx6RvDx?usp=sharing)
### Example Text2Image
```python
# !pip install diffusers==0.3.0
from diffusers import StableDiffusionPipeline
# using the 60,000 steps checkpoint
pipe = StableDiffusionPipeline.from_pretrained("naclbit/trinart_stable_diffusion_v2", revision="diffusers-60k")
pipe.to("cuda")
image = pipe("A magical dragon flying in front of the Himalaya in manga style").images[0]
image
```
If you want to run the pipeline faster or on a different hardware, please have a look at the [optimization docs](https://huggingface.co/docs/diffusers/optimization/fp16).
### Example Image2Image
```python
# !pip install diffusers==0.3.0
from diffusers import StableDiffusionImg2ImgPipeline
import requests
from PIL import Image
from io import BytesIO
url = "https://scitechdaily.com/images/Dog-Park.jpg"
response = requests.get(url)
init_image = Image.open(BytesIO(response.content)).convert("RGB")
init_image = init_image.resize((768, 512))
# using the 115,000 steps checkpoint
pipe = StableDiffusionImg2ImgPipeline.from_pretrained("naclbit/trinart_stable_diffusion_v2", revision="diffusers-115k")
pipe.to("cuda")
images = pipe(prompt="Manga drawing of Brad Pitt", init_image=init_image, strength=0.75, guidance_scale=7.5).images
image
```
If you want to run the pipeline faster or on a different hardware, please have a look at the [optimization docs](https://huggingface.co/docs/diffusers/optimization/fp16).
## Stable Diffusion TrinArt/Trin-sama AI finetune v2
trinart_stable_diffusion is a SD model finetuned by about 40,000 assorted high resolution manga/anime-style pictures for 8 epochs. This is the same model running on Twitter bot @trinsama (https://twitter.com/trinsama)
Twitterボット「とりんさまAI」@trinsama (https://twitter.com/trinsama) で使用しているSDのファインチューン済モデルです。一定のルールで選別された約4万枚のアニメ・マンガスタイルの高解像度画像を用いて約8エポックの訓練を行いました。
## Version 2
V2 checkpoint uses dropouts, 10,000 more images and a new tagging strategy and trained longer to improve results while retaining the original aesthetics.
バージョン2は画像を1万枚追加したほか、ドロップアウトの適用、タグ付けの改善とより長いトレーニング時間により、SDのスタイルを保ったまま出力内容の改善を目指しています。
## Three flavors
Step 115000/95000 checkpoints were trained further, but you may use step 60000 checkpoint instead if style nudging is too much.
ステップ115000/95000のチェックポイントでスタイルが変わりすぎると感じる場合は、ステップ60000のチェックポイントを使用してみてください。
#### img2img
If you want to run **latent-diffusion**'s stock ddim img2img script with this model, **use_ema** must be set to False.
**latent-diffusion** のscriptsフォルダに入っているddim img2imgをこのモデルで動かす場合、use_emaはFalseにする必要があります。
#### Hardware
- 8xNVIDIA A100 40GB
#### Training Info
- Custom dataset loader with augmentations: XFlip, center crop and aspect-ratio locked scaling
- LR: 1.0e-5
- 10% dropouts
#### Examples
Each images were diffused using K. Crowson's k-lms (from k-diffusion repo) method for 50 steps.
#### Credits
- Sta, AI Novelist Dev (https://ai-novel.com/) @ Bit192, Inc.
- Stable Diffusion - Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bjorn
#### License
CreativeML OpenRAIL-M | 4,752 | [
[
-0.039276123046875,
-0.048431396484375,
0.026824951171875,
0.017120361328125,
-0.0299072265625,
0.00843048095703125,
0.0098876953125,
-0.020782470703125,
0.02850341796875,
0.0225982666015625,
-0.03753662109375,
-0.0423583984375,
-0.04217529296875,
-0.007476806640625,
0.0016584396362304688,
0.06500244140625,
-0.0188446044921875,
0.0015687942504882812,
0.01611328125,
0.0006551742553710938,
-0.0170135498046875,
-0.003673553466796875,
-0.0831298828125,
-0.0305633544921875,
0.036224365234375,
0.0052947998046875,
0.045440673828125,
0.03912353515625,
0.029083251953125,
0.017181396484375,
-0.036865234375,
-0.00214385986328125,
-0.031280517578125,
-0.01520538330078125,
0.001148223876953125,
-0.021697998046875,
-0.0302886962890625,
-0.0031795501708984375,
0.057098388671875,
0.01342010498046875,
-0.00933074951171875,
0.0036640167236328125,
0.004364013671875,
0.05908203125,
-0.03387451171875,
0.00035953521728515625,
0.0013103485107421875,
0.00826263427734375,
-0.0263671875,
0.0069427490234375,
-0.00830841064453125,
-0.017669677734375,
-0.001369476318359375,
-0.0653076171875,
0.017974853515625,
-0.0003292560577392578,
0.0916748046875,
0.0291595458984375,
-0.0185699462890625,
0.0008287429809570312,
-0.03912353515625,
0.05560302734375,
-0.048675537109375,
0.0084228515625,
0.01189422607421875,
0.0175323486328125,
-0.005458831787109375,
-0.06549072265625,
-0.03961181640625,
0.0128936767578125,
0.00206756591796875,
0.024261474609375,
-0.0212860107421875,
-0.0043792724609375,
0.026397705078125,
0.0262451171875,
-0.049407958984375,
-0.0108795166015625,
-0.0301361083984375,
-0.0044403076171875,
0.053070068359375,
0.005657196044921875,
0.03936767578125,
-0.01200103759765625,
-0.0416259765625,
-0.017578125,
-0.0211029052734375,
-0.002979278564453125,
0.0116119384765625,
-0.0213470458984375,
-0.043548583984375,
0.0291290283203125,
0.00083160400390625,
0.0253448486328125,
0.0198822021484375,
-0.007251739501953125,
0.0350341796875,
-0.034271240234375,
-0.0236968994140625,
-0.0250244140625,
0.06982421875,
0.041961669921875,
0.00710296630859375,
0.01197052001953125,
-0.0024089813232421875,
-0.01617431640625,
-0.003269195556640625,
-0.09552001953125,
-0.0250091552734375,
0.032989501953125,
-0.04498291015625,
-0.030731201171875,
-0.0237884521484375,
-0.07891845703125,
-0.02099609375,
0.0173797607421875,
0.0322265625,
-0.04486083984375,
-0.04913330078125,
0.0115203857421875,
-0.034393310546875,
0.0230712890625,
0.0301361083984375,
-0.0311279296875,
0.00868988037109375,
0.02423095703125,
0.098388671875,
-0.003093719482421875,
-0.0006794929504394531,
-0.004322052001953125,
0.0115203857421875,
-0.02728271484375,
0.057342529296875,
-0.030120849609375,
-0.033477783203125,
-0.0269012451171875,
0.00444793701171875,
-0.0212554931640625,
-0.03192138671875,
0.043914794921875,
-0.042694091796875,
0.0225372314453125,
-0.0174407958984375,
-0.035308837890625,
-0.0380859375,
-0.0011472702026367188,
-0.0418701171875,
0.07684326171875,
0.028411865234375,
-0.073486328125,
0.00518035888671875,
-0.0728759765625,
0.0004074573516845703,
0.0028629302978515625,
0.0103912353515625,
-0.0478515625,
-0.0203399658203125,
-0.00004875659942626953,
0.034881591796875,
-0.016082763671875,
0.00421905517578125,
-0.040283203125,
-0.0269622802734375,
-0.00238800048828125,
-0.001827239990234375,
0.09814453125,
0.027801513671875,
-0.035247802734375,
-0.0138092041015625,
-0.04510498046875,
-0.002124786376953125,
0.0236663818359375,
-0.0153961181640625,
-0.0155792236328125,
-0.039276123046875,
0.00992584228515625,
0.0283660888671875,
0.011322021484375,
-0.0438232421875,
0.01561737060546875,
-0.023956298828125,
0.0249481201171875,
0.054107666015625,
0.03558349609375,
0.0452880859375,
-0.035491943359375,
0.04864501953125,
0.0181427001953125,
0.012939453125,
0.0031108856201171875,
-0.056915283203125,
-0.046722412109375,
-0.04327392578125,
0.01934814453125,
0.039764404296875,
-0.060821533203125,
0.01678466796875,
0.0027713775634765625,
-0.056488037109375,
-0.0275726318359375,
0.00007033348083496094,
0.0318603515625,
0.0361328125,
0.0189666748046875,
-0.038421630859375,
-0.0207672119140625,
-0.050262451171875,
-0.003498077392578125,
-0.0091552734375,
-0.0018463134765625,
0.027587890625,
0.0404052734375,
-0.0211639404296875,
0.04766845703125,
-0.05316162109375,
-0.0313720703125,
-0.0006594657897949219,
0.01285552978515625,
0.0244293212890625,
0.049713134765625,
0.06597900390625,
-0.0631103515625,
-0.069580078125,
-0.004444122314453125,
-0.061553955078125,
-0.006824493408203125,
-0.00342559814453125,
-0.039276123046875,
0.028350830078125,
0.0303497314453125,
-0.057891845703125,
0.04107666015625,
0.035247802734375,
-0.036865234375,
0.03436279296875,
-0.021453857421875,
0.00626373291015625,
-0.10205078125,
0.0265655517578125,
0.0248870849609375,
-0.0185394287109375,
-0.04888916015625,
0.00893402099609375,
-0.0087432861328125,
0.002056121826171875,
-0.04583740234375,
0.0767822265625,
-0.04205322265625,
0.023773193359375,
-0.0132904052734375,
-0.0092315673828125,
0.01029205322265625,
0.046905517578125,
0.004024505615234375,
0.04388427734375,
0.05517578125,
-0.039337158203125,
0.0161285400390625,
0.0165252685546875,
-0.015655517578125,
0.050628662109375,
-0.065673828125,
0.01007843017578125,
-0.0214385986328125,
0.03790283203125,
-0.09002685546875,
-0.02154541015625,
0.061798095703125,
-0.0335693359375,
0.043304443359375,
-0.0290679931640625,
-0.03955078125,
-0.0271453857421875,
-0.032989501953125,
0.036773681640625,
0.061553955078125,
-0.026153564453125,
0.053497314453125,
0.01097869873046875,
0.00878143310546875,
-0.04229736328125,
-0.053619384765625,
-0.009307861328125,
-0.042205810546875,
-0.06231689453125,
0.039520263671875,
-0.02191162109375,
-0.00868988037109375,
0.0015106201171875,
0.0105438232421875,
-0.01200103759765625,
-0.000980377197265625,
0.02178955078125,
0.0206756591796875,
-0.0019931793212890625,
-0.02252197265625,
0.0094757080078125,
-0.018096923828125,
-0.001628875732421875,
-0.0146484375,
0.048431396484375,
-0.002155303955078125,
-0.00652313232421875,
-0.074462890625,
0.0115203857421875,
0.037322998046875,
0.01247406005859375,
0.05859375,
0.08514404296875,
-0.0281219482421875,
0.01070404052734375,
-0.0273590087890625,
-0.0017852783203125,
-0.035980224609375,
0.00311279296875,
-0.0228729248046875,
-0.0270233154296875,
0.053924560546875,
0.00086212158203125,
0.00933837890625,
0.056610107421875,
0.038299560546875,
-0.023834228515625,
0.07147216796875,
0.0233917236328125,
0.01837158203125,
0.048492431640625,
-0.075927734375,
-0.0014333724975585938,
-0.0870361328125,
-0.0285797119140625,
-0.0212860107421875,
-0.039825439453125,
-0.034271240234375,
-0.0557861328125,
0.0338134765625,
0.021697998046875,
-0.036834716796875,
0.0079193115234375,
-0.03289794921875,
0.03338623046875,
0.0206451416015625,
0.0158233642578125,
0.0048828125,
0.015960693359375,
-0.0167999267578125,
-0.0047454833984375,
-0.04119873046875,
-0.0262603759765625,
0.06878662109375,
0.034515380859375,
0.052001953125,
0.00759124755859375,
0.0634765625,
0.0107879638671875,
0.01336669921875,
-0.024383544921875,
0.033172607421875,
-0.01352691650390625,
-0.060394287109375,
-0.0019369125366210938,
-0.0211029052734375,
-0.056915283203125,
0.0249176025390625,
0.0024127960205078125,
-0.04241943359375,
0.0335693359375,
-0.004238128662109375,
-0.026947021484375,
0.042816162109375,
-0.06610107421875,
0.057586669921875,
0.0060577392578125,
-0.0484619140625,
-0.00415802001953125,
-0.058990478515625,
0.0252838134765625,
0.0011568069458007812,
0.00905609130859375,
-0.004085540771484375,
-0.0189208984375,
0.06927490234375,
-0.03338623046875,
0.0650634765625,
-0.042022705078125,
-0.0015811920166015625,
0.0302581787109375,
0.006591796875,
0.0217437744140625,
0.00917816162109375,
-0.01393890380859375,
0.02587890625,
0.01009368896484375,
-0.04766845703125,
-0.028533935546875,
0.044677734375,
-0.061279296875,
-0.01317596435546875,
-0.04486083984375,
-0.024261474609375,
0.0304107666015625,
0.0258636474609375,
0.04962158203125,
0.0149078369140625,
0.0024852752685546875,
-0.002532958984375,
0.05426025390625,
-0.0174102783203125,
0.057098388671875,
0.00962066650390625,
-0.0203399658203125,
-0.048248291015625,
0.064697265625,
0.0112762451171875,
0.03485107421875,
0.0083770751953125,
0.0219573974609375,
-0.0251007080078125,
-0.0340576171875,
-0.04913330078125,
0.05010986328125,
-0.04229736328125,
-0.019622802734375,
-0.0609130859375,
-0.0263671875,
-0.0182952880859375,
-0.01517486572265625,
-0.0307769775390625,
-0.038726806640625,
-0.0631103515625,
0.02789306640625,
0.04095458984375,
0.044830322265625,
-0.0256195068359375,
0.037628173828125,
-0.031890869140625,
0.022064208984375,
-0.00572967529296875,
0.0287322998046875,
0.0139007568359375,
-0.05816650390625,
-0.01062774658203125,
0.0159912109375,
-0.041473388671875,
-0.0543212890625,
0.037567138671875,
0.0164947509765625,
0.02484130859375,
0.0379638671875,
-0.01235198974609375,
0.056610107421875,
-0.0115814208984375,
0.064697265625,
0.01357269287109375,
-0.0426025390625,
0.03387451171875,
-0.036224365234375,
0.0173187255859375,
0.02215576171875,
0.04998779296875,
-0.038970947265625,
-0.016021728515625,
-0.071044921875,
-0.0518798828125,
0.05902099609375,
0.029205322265625,
-0.0018053054809570312,
0.0202484130859375,
0.047393798828125,
-0.01422882080078125,
-0.004215240478515625,
-0.0462646484375,
-0.0390625,
-0.00421142578125,
0.0057525634765625,
0.0011348724365234375,
-0.0098724365234375,
-0.00653839111328125,
-0.0308990478515625,
0.07501220703125,
0.003078460693359375,
0.033843994140625,
0.02532958984375,
0.015533447265625,
-0.006656646728515625,
-0.0198822021484375,
0.0278778076171875,
0.022552490234375,
-0.030517578125,
-0.020050048828125,
-0.009185791015625,
-0.04290771484375,
0.006305694580078125,
0.003536224365234375,
-0.03253173828125,
0.013336181640625,
0.0037975311279296875,
0.057098388671875,
-0.007495880126953125,
-0.007038116455078125,
0.03778076171875,
-0.01983642578125,
-0.0285491943359375,
-0.0209197998046875,
0.014556884765625,
0.0169830322265625,
0.038116455078125,
0.0172271728515625,
0.0267791748046875,
0.0147247314453125,
-0.0106964111328125,
0.000995635986328125,
0.03350830078125,
-0.011871337890625,
-0.01971435546875,
0.061004638671875,
-0.006114959716796875,
-0.01416778564453125,
0.029388427734375,
-0.041534423828125,
-0.031585693359375,
0.056732177734375,
0.0367431640625,
0.08154296875,
-0.02069091796875,
0.035888671875,
0.058349609375,
-0.0007729530334472656,
-0.0181427001953125,
0.0281982421875,
0.00823211669921875,
-0.028106689453125,
-0.00595855712890625,
-0.050018310546875,
-0.005218505859375,
0.0030059814453125,
-0.0251312255859375,
0.038421630859375,
-0.0570068359375,
-0.01428985595703125,
0.0038738250732421875,
0.0206756591796875,
-0.0400390625,
0.01255035400390625,
0.0022907257080078125,
0.072021484375,
-0.0692138671875,
0.045623779296875,
0.04046630859375,
-0.04888916015625,
-0.050628662109375,
0.0092620849609375,
-0.002399444580078125,
-0.053619384765625,
0.03106689453125,
0.00847625732421875,
-0.0008087158203125,
0.015777587890625,
-0.05328369140625,
-0.06488037109375,
0.1112060546875,
0.01678466796875,
-0.00814056396484375,
-0.01444244384765625,
-0.0188446044921875,
0.057708740234375,
-0.034454345703125,
0.039031982421875,
0.0240020751953125,
0.02783203125,
0.042999267578125,
-0.047027587890625,
0.012237548828125,
-0.0275726318359375,
0.031768798828125,
-0.004474639892578125,
-0.089111328125,
0.06097412109375,
-0.0302886962890625,
-0.03424072265625,
0.0196685791015625,
0.035675048828125,
0.0269317626953125,
0.03424072265625,
0.0301971435546875,
0.07867431640625,
0.04827880859375,
-0.0011072158813476562,
0.0916748046875,
-0.006381988525390625,
0.04144287109375,
0.04833984375,
0.0031909942626953125,
0.043548583984375,
0.02587890625,
-0.0174102783203125,
0.061981201171875,
0.07525634765625,
-0.0009608268737792969,
0.042694091796875,
0.00710296630859375,
-0.025787353515625,
-0.0011224746704101562,
-0.0163116455078125,
-0.04119873046875,
-0.0011348724365234375,
0.015869140625,
-0.0270233154296875,
-0.0209808349609375,
-0.0035419464111328125,
0.0233917236328125,
-0.0303802490234375,
-0.00833892822265625,
0.051971435546875,
0.0091094970703125,
-0.02606201171875,
0.0762939453125,
-0.0012006759643554688,
0.05926513671875,
-0.02777099609375,
-0.004680633544921875,
-0.02374267578125,
0.005458831787109375,
-0.0313720703125,
-0.08489990234375,
0.027923583984375,
-0.0049591064453125,
0.0015001296997070312,
-0.0232696533203125,
0.0521240234375,
-0.0188446044921875,
-0.042694091796875,
0.03228759765625,
0.007694244384765625,
0.046356201171875,
-0.0032138824462890625,
-0.076416015625,
0.016571044921875,
0.011474609375,
-0.0247344970703125,
0.0104522705078125,
0.0202178955078125,
0.0138397216796875,
0.049072265625,
0.037567138671875,
0.0009565353393554688,
0.01238250732421875,
-0.0078582763671875,
0.0655517578125,
-0.031036376953125,
-0.026519775390625,
-0.06146240234375,
0.052001953125,
-0.01519775390625,
-0.0130767822265625,
0.042144775390625,
0.05523681640625,
0.0487060546875,
-0.0007910728454589844,
0.052001953125,
-0.0284576416015625,
0.026580810546875,
-0.0323486328125,
0.0618896484375,
-0.07171630859375,
0.00994110107421875,
-0.0310211181640625,
-0.07049560546875,
-0.011322021484375,
0.05078125,
-0.002399444580078125,
0.0288543701171875,
0.0452880859375,
0.0684814453125,
-0.016204833984375,
-0.01213836669921875,
-0.006023406982421875,
0.0213470458984375,
0.0300750732421875,
0.040435791015625,
0.03387451171875,
-0.06658935546875,
0.0255279541015625,
-0.04827880859375,
-0.030670166015625,
-0.0112762451171875,
-0.059814453125,
-0.07177734375,
-0.05169677734375,
-0.050872802734375,
-0.0625,
-0.0153656005859375,
0.0364990234375,
0.057708740234375,
-0.0275115966796875,
-0.0033359527587890625,
-0.0080108642578125,
0.0156707763671875,
0.00222015380859375,
-0.0192108154296875,
0.0209808349609375,
0.004932403564453125,
-0.07122802734375,
-0.0165557861328125,
0.01525115966796875,
0.036102294921875,
-0.01471710205078125,
-0.017547607421875,
-0.03826904296875,
-0.01702880859375,
0.0184783935546875,
0.0278778076171875,
-0.04046630859375,
0.0014009475708007812,
-0.00743865966796875,
-0.0016355514526367188,
0.017333984375,
0.0189208984375,
-0.029754638671875,
0.0282440185546875,
0.05291748046875,
0.0204315185546875,
0.07122802734375,
0.0021228790283203125,
0.006381988525390625,
-0.03875732421875,
0.00548553466796875,
0.00428009033203125,
0.010986328125,
0.0191650390625,
-0.032135009765625,
0.049072265625,
0.045562744140625,
-0.05010986328125,
-0.060546875,
-0.0022029876708984375,
-0.095703125,
-0.01422119140625,
0.104248046875,
-0.007762908935546875,
-0.042388916015625,
0.005710601806640625,
-0.026458740234375,
0.027191162109375,
-0.03961181640625,
0.038482666015625,
0.05560302734375,
-0.0156097412109375,
-0.0193023681640625,
-0.0386962890625,
0.043121337890625,
0.021087646484375,
-0.0423583984375,
-0.0113677978515625,
0.0272369384765625,
0.055145263671875,
0.0322265625,
0.0672607421875,
-0.0211181640625,
0.0128326416015625,
-0.002819061279296875,
0.0035457611083984375,
0.0170135498046875,
-0.004161834716796875,
-0.041015625,
-0.0137939453125,
-0.021026611328125,
-0.0204315185546875
]
] |
hoskinson-center/proofGPT-v0.1-6.7B | 2023-02-15T19:09:41.000Z | [
"transformers",
"pytorch",
"gpt_neox",
"text-generation",
"text generation",
"causal-lm",
"en",
"dataset:hoskinson-center/proof-pile",
"license:mit",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | hoskinson-center | null | null | hoskinson-center/proofGPT-v0.1-6.7B | 9 | 7,673 | transformers | 2023-02-04T23:35:52 | ---
language:
- en
tags:
- text generation
- pytorch
- causal-lm
- gpt_neox
license: mit
datasets:
- hoskinson-center/proof-pile
---
# ProofGPT-v0.1
# Model Description
ProofGPT-v0.1 is a 6.7B parameter language model based on the GPT-NeoX architecture and trained on the [proof-pile](https://huggingface.co/datasets/hoskinson-center/proof-pile) (v1.1).
We initiailized training with pythia-6.7b weights, a precursor to the [pythia-6.9b](https://huggingface.co/EleutherAI/pythia-6.9b) model that has roughly equivalent performance.
Detailed evaluations coming soon :) | 573 | [
[
-0.037506103515625,
-0.050811767578125,
0.0379638671875,
0.006923675537109375,
-0.032684326171875,
-0.027069091796875,
0.006679534912109375,
-0.018798828125,
0.006244659423828125,
0.039093017578125,
-0.0247955322265625,
-0.022613525390625,
-0.0450439453125,
-0.02825927734375,
-0.032257080078125,
0.091552734375,
0.00879669189453125,
0.0125732421875,
-0.0019588470458984375,
-0.009490966796875,
-0.00501251220703125,
-0.05224609375,
-0.056396484375,
-0.034759521484375,
0.03570556640625,
0.01061248779296875,
0.061004638671875,
0.051513671875,
0.021331787109375,
0.010528564453125,
-0.0129547119140625,
-0.04925537109375,
-0.0226287841796875,
-0.0280609130859375,
-0.00421905517578125,
-0.01255035400390625,
-0.0418701171875,
-0.00939178466796875,
0.05084228515625,
0.031951904296875,
-0.0361328125,
0.0192413330078125,
-0.029754638671875,
0.043548583984375,
-0.022674560546875,
0.00409698486328125,
-0.0290679931640625,
-0.01268768310546875,
-0.017974853515625,
0.0125579833984375,
-0.041290283203125,
-0.0190277099609375,
0.010040283203125,
-0.0390625,
0.0235443115234375,
-0.00923919677734375,
0.079833984375,
0.01459503173828125,
-0.02203369140625,
-0.011505126953125,
-0.03326416015625,
0.03912353515625,
-0.06585693359375,
0.036865234375,
0.0291290283203125,
0.062103271484375,
-0.0014705657958984375,
-0.07232666015625,
-0.0283050537109375,
-0.021392822265625,
0.00946807861328125,
-0.0012292861938476562,
-0.007190704345703125,
0.00154876708984375,
0.04840087890625,
0.038909912109375,
-0.074462890625,
0.004940032958984375,
-0.0406494140625,
-0.009490966796875,
0.04583740234375,
-0.0167694091796875,
0.0237579345703125,
0.002826690673828125,
-0.032806396484375,
-0.012237548828125,
-0.0599365234375,
-0.00502777099609375,
0.02911376953125,
0.0021038055419921875,
-0.029541015625,
0.0204315185546875,
-0.034210205078125,
0.0548095703125,
0.0082244873046875,
0.007701873779296875,
0.007801055908203125,
0.005645751953125,
-0.02337646484375,
0.0181884765625,
0.0579833984375,
0.00007164478302001953,
0.005962371826171875,
-0.0172271728515625,
-0.0203399658203125,
0.025421142578125,
0.00965118408203125,
-0.0867919921875,
-0.0582275390625,
-0.00492095947265625,
-0.04229736328125,
-0.032135009765625,
0.01538848876953125,
-0.0227203369140625,
-0.02349853515625,
-0.0074920654296875,
0.053009033203125,
-0.04144287109375,
-0.05120849609375,
0.03717041015625,
-0.004180908203125,
0.03802490234375,
0.022796630859375,
-0.06304931640625,
0.04571533203125,
0.0504150390625,
0.054412841796875,
0.0274810791015625,
-0.0098114013671875,
-0.0299530029296875,
-0.022216796875,
0.0102691650390625,
0.06298828125,
-0.0187225341796875,
-0.006893157958984375,
-0.01091766357421875,
-0.00559234619140625,
-0.000576019287109375,
-0.022735595703125,
0.06451416015625,
-0.04766845703125,
0.032073974609375,
-0.027313232421875,
-0.06146240234375,
-0.036102294921875,
0.041290283203125,
-0.054046630859375,
0.0760498046875,
0.034332275390625,
-0.0703125,
0.0379638671875,
-0.03948974609375,
0.00238800048828125,
0.0181884765625,
0.00028514862060546875,
-0.045135498046875,
0.001720428466796875,
0.0105438232421875,
0.00611114501953125,
-0.03240966796875,
-0.004032135009765625,
-0.020599365234375,
-0.028045654296875,
-0.006977081298828125,
-0.026092529296875,
0.07391357421875,
0.00838470458984375,
-0.028472900390625,
0.021759033203125,
-0.06732177734375,
0.0156402587890625,
0.0189361572265625,
-0.0282135009765625,
-0.00612640380859375,
-0.005901336669921875,
-0.00029206275939941406,
0.0352783203125,
0.024932861328125,
-0.032257080078125,
0.038909912109375,
-0.026824951171875,
0.044464111328125,
0.028961181640625,
-0.01474761962890625,
0.0005011558532714844,
-0.0306243896484375,
0.0408935546875,
-0.00146484375,
0.011993408203125,
0.0005707740783691406,
-0.044647216796875,
-0.03826904296875,
-0.0178070068359375,
0.04718017578125,
0.0307159423828125,
-0.035308837890625,
0.0244598388671875,
-0.0163421630859375,
-0.0711669921875,
-0.0003209114074707031,
-0.01763916015625,
0.024383544921875,
0.0194549560546875,
0.0227203369140625,
-0.01413726806640625,
-0.042755126953125,
-0.06256103515625,
0.0006165504455566406,
-0.042327880859375,
0.004573822021484375,
-0.0251007080078125,
0.043792724609375,
-0.0014085769653320312,
0.057952880859375,
-0.00548553466796875,
0.0369873046875,
-0.0305938720703125,
0.007709503173828125,
0.046661376953125,
0.035491943359375,
0.059234619140625,
-0.052825927734375,
-0.032623291015625,
-0.0013399124145507812,
-0.048736572265625,
-0.01140594482421875,
0.0171051025390625,
-0.005100250244140625,
0.0010118484497070312,
0.01534271240234375,
-0.0596923828125,
0.04345703125,
0.06060791015625,
-0.062103271484375,
0.06866455078125,
-0.016082763671875,
-0.007213592529296875,
-0.07659912109375,
-0.013214111328125,
0.003345489501953125,
-0.0372314453125,
-0.0291900634765625,
0.017974853515625,
0.0160980224609375,
0.007625579833984375,
-0.040191650390625,
0.040130615234375,
-0.039764404296875,
0.0003323554992675781,
-0.0276031494140625,
-0.00970458984375,
-0.0238037109375,
0.032196044921875,
0.0099334716796875,
0.045440673828125,
0.07318115234375,
-0.0537109375,
0.041656494140625,
0.00936126708984375,
-0.00653076171875,
0.0246124267578125,
-0.08087158203125,
0.01953125,
-0.00013566017150878906,
0.006847381591796875,
-0.0543212890625,
-0.002674102783203125,
0.0196533203125,
-0.0282135009765625,
0.0004296302795410156,
-0.049652099609375,
-0.0439453125,
-0.0260772705078125,
-0.010650634765625,
0.04388427734375,
0.058563232421875,
-0.0533447265625,
0.035888671875,
0.0157470703125,
0.007091522216796875,
-0.01256561279296875,
-0.019622802734375,
-0.01322174072265625,
-0.025970458984375,
-0.061126708984375,
0.004779815673828125,
-0.00951385498046875,
-0.02813720703125,
-0.0173797607421875,
-0.0088958740234375,
-0.0029392242431640625,
-0.00921630859375,
0.021331787109375,
0.00704193115234375,
-0.0309295654296875,
-0.0100860595703125,
-0.007602691650390625,
-0.02252197265625,
0.01540374755859375,
-0.051513671875,
0.060821533203125,
-0.0350341796875,
-0.018585205078125,
-0.01364898681640625,
0.0005545616149902344,
0.036041259765625,
-0.00411224365234375,
0.06341552734375,
0.035186767578125,
-0.044586181640625,
0.002056121826171875,
-0.0277862548828125,
-0.0183258056640625,
-0.0308837890625,
0.0225067138671875,
-0.037841796875,
-0.053680419921875,
0.0513916015625,
0.00557708740234375,
-0.00623321533203125,
0.046478271484375,
0.050384521484375,
0.03826904296875,
0.07781982421875,
0.03900146484375,
-0.0120391845703125,
0.0228118896484375,
-0.0180206298828125,
0.00466156005859375,
-0.046173095703125,
0.01476287841796875,
-0.03692626953125,
-0.00887298583984375,
-0.05328369140625,
-0.0196380615234375,
0.0168304443359375,
0.04803466796875,
-0.047607421875,
0.05609130859375,
-0.039703369140625,
0.0396728515625,
0.05450439453125,
0.01983642578125,
0.012939453125,
-0.01094818115234375,
-0.0034046173095703125,
0.00653076171875,
-0.07049560546875,
-0.0484619140625,
0.0753173828125,
0.04522705078125,
0.048309326171875,
0.027618408203125,
0.040008544921875,
-0.00811767578125,
0.040863037109375,
-0.048828125,
0.044891357421875,
0.010833740234375,
-0.05560302734375,
-0.01617431640625,
-0.033660888671875,
-0.0467529296875,
0.02630615234375,
-0.004192352294921875,
-0.064697265625,
-0.0214691162109375,
0.020538330078125,
-0.00518798828125,
0.033538818359375,
-0.0545654296875,
0.09619140625,
-0.0245513916015625,
-0.0134124755859375,
-0.0175323486328125,
-0.033782958984375,
0.017425537109375,
0.04718017578125,
-0.0100250244140625,
-0.00572967529296875,
0.037445068359375,
0.052001953125,
-0.042388916015625,
0.032257080078125,
-0.03173828125,
-0.00411224365234375,
0.0266571044921875,
0.0173187255859375,
0.056640625,
0.0300445556640625,
-0.01081085205078125,
0.01422882080078125,
0.0195770263671875,
-0.048248291015625,
-0.0251007080078125,
0.04595947265625,
-0.05352783203125,
-0.0155487060546875,
-0.0477294921875,
-0.03363037109375,
0.0002815723419189453,
0.031707763671875,
0.031341552734375,
0.0345458984375,
-0.024993896484375,
0.006900787353515625,
0.0762939453125,
0.003787994384765625,
0.035186767578125,
0.03594970703125,
-0.041015625,
-0.0270538330078125,
0.05975341796875,
-0.005603790283203125,
0.0257110595703125,
0.0296478271484375,
0.040863037109375,
-0.0264129638671875,
-0.05474853515625,
-0.046051025390625,
0.0254058837890625,
-0.042694091796875,
-0.000331878662109375,
-0.0291290283203125,
-0.027496337890625,
-0.044708251953125,
0.0100250244140625,
-0.03485107421875,
-0.041534423828125,
0.0232086181640625,
-0.0173797607421875,
0.03814697265625,
0.056060791015625,
0.0155029296875,
0.0416259765625,
-0.055267333984375,
0.00982666015625,
0.0303497314453125,
0.042022705078125,
-0.041107177734375,
-0.05950927734375,
-0.0254058837890625,
-0.004150390625,
-0.0168914794921875,
-0.07489013671875,
0.0262451171875,
-0.003841400146484375,
0.039642333984375,
0.036376953125,
-0.0259246826171875,
0.00849151611328125,
-0.018798828125,
0.05682373046875,
0.00734710693359375,
-0.06298828125,
0.034515380859375,
-0.05670166015625,
0.0294952392578125,
0.0197906494140625,
0.0179443359375,
-0.036041259765625,
0.00421905517578125,
-0.0767822265625,
-0.0562744140625,
0.033905029296875,
0.0227203369140625,
-0.0283050537109375,
0.023590087890625,
0.03570556640625,
0.0176544189453125,
0.0160980224609375,
-0.07080078125,
-0.00455474853515625,
-0.043212890625,
-0.0240325927734375,
0.0205841064453125,
-0.0273590087890625,
0.004425048828125,
-0.0225677490234375,
0.062103271484375,
-0.01161956787109375,
0.01512908935546875,
0.0033855438232421875,
-0.01528167724609375,
0.01561737060546875,
0.01263427734375,
0.05352783203125,
0.0743408203125,
-0.02362060546875,
0.007373809814453125,
0.0034999847412109375,
-0.045135498046875,
-0.005474090576171875,
0.0325927734375,
-0.035064697265625,
-0.001140594482421875,
0.0129547119140625,
0.07989501953125,
-0.00202178955078125,
-0.033660888671875,
0.0183868408203125,
-0.01885986328125,
-0.0360107421875,
-0.03948974609375,
0.00553131103515625,
0.0019502639770507812,
0.0010080337524414062,
0.00289154052734375,
-0.0163421630859375,
0.01129150390625,
-0.0227813720703125,
0.01898193359375,
0.0171966552734375,
-0.033477783203125,
-0.0450439453125,
0.050628662109375,
-0.0189361572265625,
-0.015228271484375,
0.03546142578125,
-0.043975830078125,
-0.04290771484375,
0.0239410400390625,
0.052215576171875,
0.05950927734375,
-0.00852203369140625,
0.031402587890625,
0.0491943359375,
0.031280517578125,
-0.0253753662109375,
0.0243072509765625,
0.0008382797241210938,
-0.053009033203125,
-0.0238037109375,
-0.056060791015625,
-0.03204345703125,
0.00168609619140625,
-0.033905029296875,
0.0240631103515625,
-0.0618896484375,
0.006893157958984375,
-0.020416259765625,
0.0009984970092773438,
-0.0284576416015625,
0.030731201171875,
0.0096435546875,
0.08251953125,
-0.053863525390625,
0.0728759765625,
0.077880859375,
-0.038604736328125,
-0.0838623046875,
-0.0206298828125,
-0.00714874267578125,
-0.042816162109375,
0.0231170654296875,
0.00708770751953125,
0.01617431640625,
0.016448974609375,
-0.05560302734375,
-0.060516357421875,
0.073486328125,
0.04791259765625,
-0.040435791015625,
-0.001514434814453125,
-0.006916046142578125,
0.02911376953125,
-0.01140594482421875,
0.02386474609375,
0.03778076171875,
0.030059814453125,
0.0008282661437988281,
-0.08111572265625,
0.004314422607421875,
-0.039276123046875,
0.0045318603515625,
0.0323486328125,
-0.04095458984375,
0.07550048828125,
0.00855255126953125,
0.004421234130859375,
0.0037136077880859375,
0.06256103515625,
0.0423583984375,
0.0019626617431640625,
0.04638671875,
0.0867919921875,
0.033203125,
-0.028228759765625,
0.0843505859375,
-0.0300750732421875,
0.05194091796875,
0.0755615234375,
-0.002864837646484375,
0.038055419921875,
0.0271453857421875,
-0.02239990234375,
0.04888916015625,
0.0435791015625,
0.016998291015625,
0.022552490234375,
0.026092529296875,
-0.0230255126953125,
-0.03497314453125,
0.020111083984375,
-0.048583984375,
0.00824737548828125,
0.01024627685546875,
-0.014678955078125,
-0.0183868408203125,
-0.02685546875,
-0.00006538629531860352,
-0.0074005126953125,
-0.03289794921875,
0.040252685546875,
-0.002498626708984375,
-0.03277587890625,
0.031494140625,
0.0056915283203125,
0.041259765625,
-0.052886962890625,
-0.0102996826171875,
0.0082550048828125,
0.022918701171875,
0.00229644775390625,
-0.04815673828125,
0.004528045654296875,
-0.0252532958984375,
-0.0187225341796875,
-0.0092926025390625,
0.07305908203125,
-0.0254364013671875,
-0.048248291015625,
0.0036106109619140625,
0.0161895751953125,
0.01125335693359375,
-0.0389404296875,
-0.055145263671875,
0.003498077392578125,
-0.035614013671875,
-0.036956787109375,
0.00804901123046875,
0.03369140625,
-0.01444244384765625,
0.046478271484375,
0.0343017578125,
0.01540374755859375,
0.0204925537109375,
0.03204345703125,
0.05364990234375,
-0.046600341796875,
-0.06390380859375,
-0.05120849609375,
0.0474853515625,
-0.0034618377685546875,
-0.033660888671875,
0.06146240234375,
0.04400634765625,
0.0657958984375,
-0.013702392578125,
0.0638427734375,
-0.0406494140625,
0.0400390625,
-0.04583740234375,
0.04010009765625,
-0.0050506591796875,
-0.0100555419921875,
-0.01401519775390625,
-0.06640625,
-0.00913238525390625,
0.06695556640625,
-0.02545166015625,
0.0247039794921875,
0.050384521484375,
0.052154541015625,
-0.005252838134765625,
0.01020050048828125,
0.0248870849609375,
0.0200958251953125,
0.0340576171875,
0.055755615234375,
0.04669189453125,
-0.077392578125,
0.024658203125,
-0.0239715576171875,
-0.019561767578125,
-0.002925872802734375,
-0.071044921875,
-0.06781005859375,
-0.0277099609375,
-0.032623291015625,
-0.05340576171875,
0.006153106689453125,
0.06689453125,
0.072021484375,
-0.05389404296875,
-0.013275146484375,
-0.040435791015625,
-0.005237579345703125,
-0.0283355712890625,
-0.0131683349609375,
0.0193328857421875,
0.0178680419921875,
-0.027618408203125,
0.0160675048828125,
0.01393890380859375,
0.005039215087890625,
-0.030517578125,
-0.0362548828125,
-0.0179443359375,
-0.0154266357421875,
0.0298614501953125,
0.033843994140625,
-0.023193359375,
-0.029754638671875,
-0.0187835693359375,
0.0019397735595703125,
0.0018978118896484375,
0.063232421875,
-0.05157470703125,
-0.003265380859375,
0.062042236328125,
0.0268402099609375,
0.04437255859375,
0.0058746337890625,
0.04449462890625,
-0.03485107421875,
0.03717041015625,
0.012603759765625,
0.0477294921875,
0.0278472900390625,
-0.01371002197265625,
0.038055419921875,
0.039154052734375,
-0.058258056640625,
-0.038299560546875,
0.02587890625,
-0.08642578125,
0.01435089111328125,
0.07208251953125,
-0.0008487701416015625,
-0.023406982421875,
0.01287078857421875,
-0.0292205810546875,
0.016998291015625,
-0.007068634033203125,
0.05059814453125,
0.05438232421875,
0.00982666015625,
-0.042816162109375,
-0.049774169921875,
0.02935791015625,
0.0238800048828125,
-0.0533447265625,
-0.0021991729736328125,
0.034637451171875,
0.03839111328125,
0.006977081298828125,
0.059295654296875,
-0.024139404296875,
0.048583984375,
0.017425537109375,
0.0172119140625,
-0.010650634765625,
-0.02490234375,
-0.013275146484375,
0.016021728515625,
0.0270538330078125,
0.004863739013671875
]
] |
intfloat/e5-base | 2023-08-07T04:59:19.000Z | [
"sentence-transformers",
"pytorch",
"safetensors",
"bert",
"mteb",
"Sentence Transformers",
"sentence-similarity",
"en",
"arxiv:2212.03533",
"arxiv:2104.08663",
"arxiv:2210.07316",
"license:mit",
"model-index",
"endpoints_compatible",
"has_space",
"region:us"
] | sentence-similarity | intfloat | null | null | intfloat/e5-base | 15 | 7,664 | sentence-transformers | 2022-12-26T05:58:05 | ---
tags:
- mteb
- Sentence Transformers
- sentence-similarity
- sentence-transformers
model-index:
- name: e5-base
results:
- task:
type: Classification
dataset:
type: mteb/amazon_counterfactual
name: MTEB AmazonCounterfactualClassification (en)
config: en
split: test
revision: e8379541af4e31359cca9fbcf4b00f2671dba205
metrics:
- type: accuracy
value: 79.71641791044777
- type: ap
value: 44.15426065428253
- type: f1
value: 73.89474407693241
- task:
type: Classification
dataset:
type: mteb/amazon_polarity
name: MTEB AmazonPolarityClassification
config: default
split: test
revision: e2d317d38cd51312af73b3d32a06d1a08b442046
metrics:
- type: accuracy
value: 87.9649
- type: ap
value: 84.10171551915973
- type: f1
value: 87.94148377827356
- task:
type: Classification
dataset:
type: mteb/amazon_reviews_multi
name: MTEB AmazonReviewsClassification (en)
config: en
split: test
revision: 1399c76144fd37290681b995c656ef9b2e06e26d
metrics:
- type: accuracy
value: 42.645999999999994
- type: f1
value: 42.230574673549
- task:
type: Retrieval
dataset:
type: arguana
name: MTEB ArguAna
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 26.814
- type: map_at_10
value: 42.681999999999995
- type: map_at_100
value: 43.714
- type: map_at_1000
value: 43.724000000000004
- type: map_at_3
value: 38.11
- type: map_at_5
value: 40.666999999999994
- type: mrr_at_1
value: 27.168999999999997
- type: mrr_at_10
value: 42.84
- type: mrr_at_100
value: 43.864
- type: mrr_at_1000
value: 43.875
- type: mrr_at_3
value: 38.193
- type: mrr_at_5
value: 40.793
- type: ndcg_at_1
value: 26.814
- type: ndcg_at_10
value: 51.410999999999994
- type: ndcg_at_100
value: 55.713
- type: ndcg_at_1000
value: 55.957
- type: ndcg_at_3
value: 41.955
- type: ndcg_at_5
value: 46.558
- type: precision_at_1
value: 26.814
- type: precision_at_10
value: 7.922999999999999
- type: precision_at_100
value: 0.9780000000000001
- type: precision_at_1000
value: 0.1
- type: precision_at_3
value: 17.71
- type: precision_at_5
value: 12.859000000000002
- type: recall_at_1
value: 26.814
- type: recall_at_10
value: 79.232
- type: recall_at_100
value: 97.795
- type: recall_at_1000
value: 99.644
- type: recall_at_3
value: 53.129000000000005
- type: recall_at_5
value: 64.29599999999999
- task:
type: Clustering
dataset:
type: mteb/arxiv-clustering-p2p
name: MTEB ArxivClusteringP2P
config: default
split: test
revision: a122ad7f3f0291bf49cc6f4d32aa80929df69d5d
metrics:
- type: v_measure
value: 44.56933066536439
- task:
type: Clustering
dataset:
type: mteb/arxiv-clustering-s2s
name: MTEB ArxivClusteringS2S
config: default
split: test
revision: f910caf1a6075f7329cdf8c1a6135696f37dbd53
metrics:
- type: v_measure
value: 40.47647746165173
- task:
type: Reranking
dataset:
type: mteb/askubuntudupquestions-reranking
name: MTEB AskUbuntuDupQuestions
config: default
split: test
revision: 2000358ca161889fa9c082cb41daa8dcfb161a54
metrics:
- type: map
value: 59.65675531567043
- type: mrr
value: 72.95255683067317
- task:
type: STS
dataset:
type: mteb/biosses-sts
name: MTEB BIOSSES
config: default
split: test
revision: d3fb88f8f02e40887cd149695127462bbcf29b4a
metrics:
- type: cos_sim_pearson
value: 85.83147014162338
- type: cos_sim_spearman
value: 85.1031439521441
- type: euclidean_pearson
value: 83.53609085510973
- type: euclidean_spearman
value: 84.59650590202833
- type: manhattan_pearson
value: 83.14611947586386
- type: manhattan_spearman
value: 84.13384475757064
- task:
type: Classification
dataset:
type: mteb/banking77
name: MTEB Banking77Classification
config: default
split: test
revision: 0fd18e25b25c072e09e0d92ab615fda904d66300
metrics:
- type: accuracy
value: 83.32792207792208
- type: f1
value: 83.32037485050513
- task:
type: Clustering
dataset:
type: mteb/biorxiv-clustering-p2p
name: MTEB BiorxivClusteringP2P
config: default
split: test
revision: 65b79d1d13f80053f67aca9498d9402c2d9f1f40
metrics:
- type: v_measure
value: 36.18605446588703
- task:
type: Clustering
dataset:
type: mteb/biorxiv-clustering-s2s
name: MTEB BiorxivClusteringS2S
config: default
split: test
revision: 258694dd0231531bc1fd9de6ceb52a0853c6d908
metrics:
- type: v_measure
value: 32.72379130181917
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackAndroidRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 30.659
- type: map_at_10
value: 40.333999999999996
- type: map_at_100
value: 41.763
- type: map_at_1000
value: 41.894
- type: map_at_3
value: 37.561
- type: map_at_5
value: 39.084
- type: mrr_at_1
value: 37.482
- type: mrr_at_10
value: 45.736
- type: mrr_at_100
value: 46.591
- type: mrr_at_1000
value: 46.644999999999996
- type: mrr_at_3
value: 43.491
- type: mrr_at_5
value: 44.75
- type: ndcg_at_1
value: 37.482
- type: ndcg_at_10
value: 45.606
- type: ndcg_at_100
value: 51.172
- type: ndcg_at_1000
value: 53.407000000000004
- type: ndcg_at_3
value: 41.808
- type: ndcg_at_5
value: 43.449
- type: precision_at_1
value: 37.482
- type: precision_at_10
value: 8.254999999999999
- type: precision_at_100
value: 1.3719999999999999
- type: precision_at_1000
value: 0.186
- type: precision_at_3
value: 19.695
- type: precision_at_5
value: 13.847999999999999
- type: recall_at_1
value: 30.659
- type: recall_at_10
value: 55.409
- type: recall_at_100
value: 78.687
- type: recall_at_1000
value: 93.068
- type: recall_at_3
value: 43.891999999999996
- type: recall_at_5
value: 48.678
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackEnglishRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 30.977
- type: map_at_10
value: 40.296
- type: map_at_100
value: 41.453
- type: map_at_1000
value: 41.581
- type: map_at_3
value: 37.619
- type: map_at_5
value: 39.181
- type: mrr_at_1
value: 39.108
- type: mrr_at_10
value: 46.894000000000005
- type: mrr_at_100
value: 47.55
- type: mrr_at_1000
value: 47.598
- type: mrr_at_3
value: 44.766
- type: mrr_at_5
value: 46.062999999999995
- type: ndcg_at_1
value: 39.108
- type: ndcg_at_10
value: 45.717
- type: ndcg_at_100
value: 49.941
- type: ndcg_at_1000
value: 52.138
- type: ndcg_at_3
value: 42.05
- type: ndcg_at_5
value: 43.893
- type: precision_at_1
value: 39.108
- type: precision_at_10
value: 8.306
- type: precision_at_100
value: 1.3419999999999999
- type: precision_at_1000
value: 0.184
- type: precision_at_3
value: 19.979
- type: precision_at_5
value: 14.038
- type: recall_at_1
value: 30.977
- type: recall_at_10
value: 54.688
- type: recall_at_100
value: 72.556
- type: recall_at_1000
value: 86.53800000000001
- type: recall_at_3
value: 43.388
- type: recall_at_5
value: 48.717
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackGamingRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 39.812
- type: map_at_10
value: 50.1
- type: map_at_100
value: 51.193999999999996
- type: map_at_1000
value: 51.258
- type: map_at_3
value: 47.510999999999996
- type: map_at_5
value: 48.891
- type: mrr_at_1
value: 45.266
- type: mrr_at_10
value: 53.459999999999994
- type: mrr_at_100
value: 54.19199999999999
- type: mrr_at_1000
value: 54.228
- type: mrr_at_3
value: 51.296
- type: mrr_at_5
value: 52.495999999999995
- type: ndcg_at_1
value: 45.266
- type: ndcg_at_10
value: 55.034000000000006
- type: ndcg_at_100
value: 59.458
- type: ndcg_at_1000
value: 60.862
- type: ndcg_at_3
value: 50.52799999999999
- type: ndcg_at_5
value: 52.564
- type: precision_at_1
value: 45.266
- type: precision_at_10
value: 8.483
- type: precision_at_100
value: 1.162
- type: precision_at_1000
value: 0.133
- type: precision_at_3
value: 21.944
- type: precision_at_5
value: 14.721
- type: recall_at_1
value: 39.812
- type: recall_at_10
value: 66.36
- type: recall_at_100
value: 85.392
- type: recall_at_1000
value: 95.523
- type: recall_at_3
value: 54.127
- type: recall_at_5
value: 59.245000000000005
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackGisRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 26.186
- type: map_at_10
value: 33.18
- type: map_at_100
value: 34.052
- type: map_at_1000
value: 34.149
- type: map_at_3
value: 31.029
- type: map_at_5
value: 32.321
- type: mrr_at_1
value: 28.136
- type: mrr_at_10
value: 35.195
- type: mrr_at_100
value: 35.996
- type: mrr_at_1000
value: 36.076
- type: mrr_at_3
value: 33.051
- type: mrr_at_5
value: 34.407
- type: ndcg_at_1
value: 28.136
- type: ndcg_at_10
value: 37.275999999999996
- type: ndcg_at_100
value: 41.935
- type: ndcg_at_1000
value: 44.389
- type: ndcg_at_3
value: 33.059
- type: ndcg_at_5
value: 35.313
- type: precision_at_1
value: 28.136
- type: precision_at_10
value: 5.457999999999999
- type: precision_at_100
value: 0.826
- type: precision_at_1000
value: 0.107
- type: precision_at_3
value: 13.522
- type: precision_at_5
value: 9.424000000000001
- type: recall_at_1
value: 26.186
- type: recall_at_10
value: 47.961999999999996
- type: recall_at_100
value: 70.072
- type: recall_at_1000
value: 88.505
- type: recall_at_3
value: 36.752
- type: recall_at_5
value: 42.168
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackMathematicaRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 16.586000000000002
- type: map_at_10
value: 23.637
- type: map_at_100
value: 24.82
- type: map_at_1000
value: 24.95
- type: map_at_3
value: 21.428
- type: map_at_5
value: 22.555
- type: mrr_at_1
value: 20.771
- type: mrr_at_10
value: 27.839999999999996
- type: mrr_at_100
value: 28.887
- type: mrr_at_1000
value: 28.967
- type: mrr_at_3
value: 25.56
- type: mrr_at_5
value: 26.723000000000003
- type: ndcg_at_1
value: 20.771
- type: ndcg_at_10
value: 28.255000000000003
- type: ndcg_at_100
value: 33.886
- type: ndcg_at_1000
value: 36.963
- type: ndcg_at_3
value: 24.056
- type: ndcg_at_5
value: 25.818
- type: precision_at_1
value: 20.771
- type: precision_at_10
value: 5.1
- type: precision_at_100
value: 0.9119999999999999
- type: precision_at_1000
value: 0.132
- type: precision_at_3
value: 11.526
- type: precision_at_5
value: 8.158999999999999
- type: recall_at_1
value: 16.586000000000002
- type: recall_at_10
value: 38.456
- type: recall_at_100
value: 62.666
- type: recall_at_1000
value: 84.47
- type: recall_at_3
value: 26.765
- type: recall_at_5
value: 31.297000000000004
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackPhysicsRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 28.831
- type: map_at_10
value: 37.545
- type: map_at_100
value: 38.934999999999995
- type: map_at_1000
value: 39.044000000000004
- type: map_at_3
value: 34.601
- type: map_at_5
value: 36.302
- type: mrr_at_1
value: 34.264
- type: mrr_at_10
value: 42.569
- type: mrr_at_100
value: 43.514
- type: mrr_at_1000
value: 43.561
- type: mrr_at_3
value: 40.167
- type: mrr_at_5
value: 41.678
- type: ndcg_at_1
value: 34.264
- type: ndcg_at_10
value: 42.914
- type: ndcg_at_100
value: 48.931999999999995
- type: ndcg_at_1000
value: 51.004000000000005
- type: ndcg_at_3
value: 38.096999999999994
- type: ndcg_at_5
value: 40.509
- type: precision_at_1
value: 34.264
- type: precision_at_10
value: 7.642
- type: precision_at_100
value: 1.258
- type: precision_at_1000
value: 0.161
- type: precision_at_3
value: 17.453
- type: precision_at_5
value: 12.608
- type: recall_at_1
value: 28.831
- type: recall_at_10
value: 53.56999999999999
- type: recall_at_100
value: 79.26100000000001
- type: recall_at_1000
value: 92.862
- type: recall_at_3
value: 40.681
- type: recall_at_5
value: 46.597
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackProgrammersRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 27.461000000000002
- type: map_at_10
value: 35.885
- type: map_at_100
value: 37.039
- type: map_at_1000
value: 37.16
- type: map_at_3
value: 33.451
- type: map_at_5
value: 34.807
- type: mrr_at_1
value: 34.018
- type: mrr_at_10
value: 41.32
- type: mrr_at_100
value: 42.157
- type: mrr_at_1000
value: 42.223
- type: mrr_at_3
value: 39.288000000000004
- type: mrr_at_5
value: 40.481
- type: ndcg_at_1
value: 34.018
- type: ndcg_at_10
value: 40.821000000000005
- type: ndcg_at_100
value: 46.053
- type: ndcg_at_1000
value: 48.673
- type: ndcg_at_3
value: 36.839
- type: ndcg_at_5
value: 38.683
- type: precision_at_1
value: 34.018
- type: precision_at_10
value: 7.009
- type: precision_at_100
value: 1.123
- type: precision_at_1000
value: 0.153
- type: precision_at_3
value: 16.933
- type: precision_at_5
value: 11.826
- type: recall_at_1
value: 27.461000000000002
- type: recall_at_10
value: 50.285000000000004
- type: recall_at_100
value: 73.25500000000001
- type: recall_at_1000
value: 91.17699999999999
- type: recall_at_3
value: 39.104
- type: recall_at_5
value: 43.968
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 26.980083333333337
- type: map_at_10
value: 34.47208333333333
- type: map_at_100
value: 35.609249999999996
- type: map_at_1000
value: 35.72833333333333
- type: map_at_3
value: 32.189416666666666
- type: map_at_5
value: 33.44683333333334
- type: mrr_at_1
value: 31.731666666666662
- type: mrr_at_10
value: 38.518
- type: mrr_at_100
value: 39.38166666666667
- type: mrr_at_1000
value: 39.446999999999996
- type: mrr_at_3
value: 36.49966666666668
- type: mrr_at_5
value: 37.639916666666664
- type: ndcg_at_1
value: 31.731666666666662
- type: ndcg_at_10
value: 38.92033333333333
- type: ndcg_at_100
value: 44.01675
- type: ndcg_at_1000
value: 46.51075
- type: ndcg_at_3
value: 35.09766666666667
- type: ndcg_at_5
value: 36.842999999999996
- type: precision_at_1
value: 31.731666666666662
- type: precision_at_10
value: 6.472583333333332
- type: precision_at_100
value: 1.0665
- type: precision_at_1000
value: 0.14725000000000002
- type: precision_at_3
value: 15.659083333333331
- type: precision_at_5
value: 10.878833333333333
- type: recall_at_1
value: 26.980083333333337
- type: recall_at_10
value: 48.13925
- type: recall_at_100
value: 70.70149999999998
- type: recall_at_1000
value: 88.10775000000001
- type: recall_at_3
value: 37.30091666666667
- type: recall_at_5
value: 41.90358333333333
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackStatsRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 25.607999999999997
- type: map_at_10
value: 30.523
- type: map_at_100
value: 31.409
- type: map_at_1000
value: 31.507
- type: map_at_3
value: 28.915000000000003
- type: map_at_5
value: 29.756
- type: mrr_at_1
value: 28.681
- type: mrr_at_10
value: 33.409
- type: mrr_at_100
value: 34.241
- type: mrr_at_1000
value: 34.313
- type: mrr_at_3
value: 32.029999999999994
- type: mrr_at_5
value: 32.712
- type: ndcg_at_1
value: 28.681
- type: ndcg_at_10
value: 33.733000000000004
- type: ndcg_at_100
value: 38.32
- type: ndcg_at_1000
value: 40.937
- type: ndcg_at_3
value: 30.898999999999997
- type: ndcg_at_5
value: 32.088
- type: precision_at_1
value: 28.681
- type: precision_at_10
value: 4.968999999999999
- type: precision_at_100
value: 0.79
- type: precision_at_1000
value: 0.11
- type: precision_at_3
value: 12.73
- type: precision_at_5
value: 8.558
- type: recall_at_1
value: 25.607999999999997
- type: recall_at_10
value: 40.722
- type: recall_at_100
value: 61.956999999999994
- type: recall_at_1000
value: 81.43
- type: recall_at_3
value: 32.785
- type: recall_at_5
value: 35.855
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackTexRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 20.399
- type: map_at_10
value: 25.968000000000004
- type: map_at_100
value: 26.985999999999997
- type: map_at_1000
value: 27.105
- type: map_at_3
value: 24.215
- type: map_at_5
value: 25.157
- type: mrr_at_1
value: 24.708
- type: mrr_at_10
value: 29.971999999999998
- type: mrr_at_100
value: 30.858
- type: mrr_at_1000
value: 30.934
- type: mrr_at_3
value: 28.304000000000002
- type: mrr_at_5
value: 29.183999999999997
- type: ndcg_at_1
value: 24.708
- type: ndcg_at_10
value: 29.676000000000002
- type: ndcg_at_100
value: 34.656
- type: ndcg_at_1000
value: 37.588
- type: ndcg_at_3
value: 26.613
- type: ndcg_at_5
value: 27.919
- type: precision_at_1
value: 24.708
- type: precision_at_10
value: 5.01
- type: precision_at_100
value: 0.876
- type: precision_at_1000
value: 0.13
- type: precision_at_3
value: 11.975
- type: precision_at_5
value: 8.279
- type: recall_at_1
value: 20.399
- type: recall_at_10
value: 36.935
- type: recall_at_100
value: 59.532
- type: recall_at_1000
value: 80.58
- type: recall_at_3
value: 27.979
- type: recall_at_5
value: 31.636999999999997
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackUnixRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 27.606
- type: map_at_10
value: 34.213
- type: map_at_100
value: 35.339999999999996
- type: map_at_1000
value: 35.458
- type: map_at_3
value: 31.987
- type: map_at_5
value: 33.322
- type: mrr_at_1
value: 31.53
- type: mrr_at_10
value: 37.911
- type: mrr_at_100
value: 38.879000000000005
- type: mrr_at_1000
value: 38.956
- type: mrr_at_3
value: 35.868
- type: mrr_at_5
value: 37.047999999999995
- type: ndcg_at_1
value: 31.53
- type: ndcg_at_10
value: 38.312000000000005
- type: ndcg_at_100
value: 43.812
- type: ndcg_at_1000
value: 46.414
- type: ndcg_at_3
value: 34.319
- type: ndcg_at_5
value: 36.312
- type: precision_at_1
value: 31.53
- type: precision_at_10
value: 5.970000000000001
- type: precision_at_100
value: 0.9939999999999999
- type: precision_at_1000
value: 0.133
- type: precision_at_3
value: 14.738999999999999
- type: precision_at_5
value: 10.242999999999999
- type: recall_at_1
value: 27.606
- type: recall_at_10
value: 47.136
- type: recall_at_100
value: 71.253
- type: recall_at_1000
value: 89.39399999999999
- type: recall_at_3
value: 36.342
- type: recall_at_5
value: 41.388999999999996
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackWebmastersRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 24.855
- type: map_at_10
value: 31.963
- type: map_at_100
value: 33.371
- type: map_at_1000
value: 33.584
- type: map_at_3
value: 29.543999999999997
- type: map_at_5
value: 30.793
- type: mrr_at_1
value: 29.644
- type: mrr_at_10
value: 35.601
- type: mrr_at_100
value: 36.551
- type: mrr_at_1000
value: 36.623
- type: mrr_at_3
value: 33.399
- type: mrr_at_5
value: 34.575
- type: ndcg_at_1
value: 29.644
- type: ndcg_at_10
value: 36.521
- type: ndcg_at_100
value: 42.087
- type: ndcg_at_1000
value: 45.119
- type: ndcg_at_3
value: 32.797
- type: ndcg_at_5
value: 34.208
- type: precision_at_1
value: 29.644
- type: precision_at_10
value: 6.7
- type: precision_at_100
value: 1.374
- type: precision_at_1000
value: 0.22899999999999998
- type: precision_at_3
value: 15.152
- type: precision_at_5
value: 10.671999999999999
- type: recall_at_1
value: 24.855
- type: recall_at_10
value: 45.449
- type: recall_at_100
value: 70.921
- type: recall_at_1000
value: 90.629
- type: recall_at_3
value: 33.526
- type: recall_at_5
value: 37.848
- task:
type: Retrieval
dataset:
type: BeIR/cqadupstack
name: MTEB CQADupstackWordpressRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 24.781
- type: map_at_10
value: 30.020999999999997
- type: map_at_100
value: 30.948999999999998
- type: map_at_1000
value: 31.05
- type: map_at_3
value: 28.412
- type: map_at_5
value: 29.193
- type: mrr_at_1
value: 27.172
- type: mrr_at_10
value: 32.309
- type: mrr_at_100
value: 33.164
- type: mrr_at_1000
value: 33.239999999999995
- type: mrr_at_3
value: 30.775999999999996
- type: mrr_at_5
value: 31.562
- type: ndcg_at_1
value: 27.172
- type: ndcg_at_10
value: 33.178999999999995
- type: ndcg_at_100
value: 37.949
- type: ndcg_at_1000
value: 40.635
- type: ndcg_at_3
value: 30.107
- type: ndcg_at_5
value: 31.36
- type: precision_at_1
value: 27.172
- type: precision_at_10
value: 4.769
- type: precision_at_100
value: 0.769
- type: precision_at_1000
value: 0.109
- type: precision_at_3
value: 12.261
- type: precision_at_5
value: 8.17
- type: recall_at_1
value: 24.781
- type: recall_at_10
value: 40.699000000000005
- type: recall_at_100
value: 62.866
- type: recall_at_1000
value: 83.11699999999999
- type: recall_at_3
value: 32.269999999999996
- type: recall_at_5
value: 35.443999999999996
- task:
type: Retrieval
dataset:
type: climate-fever
name: MTEB ClimateFEVER
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 5.2139999999999995
- type: map_at_10
value: 9.986
- type: map_at_100
value: 11.343
- type: map_at_1000
value: 11.55
- type: map_at_3
value: 7.961
- type: map_at_5
value: 8.967
- type: mrr_at_1
value: 12.052
- type: mrr_at_10
value: 20.165
- type: mrr_at_100
value: 21.317
- type: mrr_at_1000
value: 21.399
- type: mrr_at_3
value: 17.079
- type: mrr_at_5
value: 18.695
- type: ndcg_at_1
value: 12.052
- type: ndcg_at_10
value: 15.375
- type: ndcg_at_100
value: 21.858
- type: ndcg_at_1000
value: 26.145000000000003
- type: ndcg_at_3
value: 11.334
- type: ndcg_at_5
value: 12.798000000000002
- type: precision_at_1
value: 12.052
- type: precision_at_10
value: 5.16
- type: precision_at_100
value: 1.206
- type: precision_at_1000
value: 0.198
- type: precision_at_3
value: 8.73
- type: precision_at_5
value: 7.114
- type: recall_at_1
value: 5.2139999999999995
- type: recall_at_10
value: 20.669999999999998
- type: recall_at_100
value: 43.901
- type: recall_at_1000
value: 68.447
- type: recall_at_3
value: 11.049000000000001
- type: recall_at_5
value: 14.652999999999999
- task:
type: Retrieval
dataset:
type: dbpedia-entity
name: MTEB DBPedia
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 8.511000000000001
- type: map_at_10
value: 19.503
- type: map_at_100
value: 27.46
- type: map_at_1000
value: 29.187
- type: map_at_3
value: 14.030999999999999
- type: map_at_5
value: 16.329
- type: mrr_at_1
value: 63.74999999999999
- type: mrr_at_10
value: 73.419
- type: mrr_at_100
value: 73.691
- type: mrr_at_1000
value: 73.697
- type: mrr_at_3
value: 71.792
- type: mrr_at_5
value: 72.979
- type: ndcg_at_1
value: 53.125
- type: ndcg_at_10
value: 41.02
- type: ndcg_at_100
value: 45.407
- type: ndcg_at_1000
value: 52.68000000000001
- type: ndcg_at_3
value: 46.088
- type: ndcg_at_5
value: 43.236000000000004
- type: precision_at_1
value: 63.74999999999999
- type: precision_at_10
value: 32.35
- type: precision_at_100
value: 10.363
- type: precision_at_1000
value: 2.18
- type: precision_at_3
value: 49.667
- type: precision_at_5
value: 41.5
- type: recall_at_1
value: 8.511000000000001
- type: recall_at_10
value: 24.851
- type: recall_at_100
value: 50.745
- type: recall_at_1000
value: 73.265
- type: recall_at_3
value: 15.716
- type: recall_at_5
value: 19.256
- task:
type: Classification
dataset:
type: mteb/emotion
name: MTEB EmotionClassification
config: default
split: test
revision: 4f58c6b202a23cf9a4da393831edf4f9183cad37
metrics:
- type: accuracy
value: 49.43500000000001
- type: f1
value: 44.56288273966374
- task:
type: Retrieval
dataset:
type: fever
name: MTEB FEVER
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 40.858
- type: map_at_10
value: 52.276
- type: map_at_100
value: 52.928
- type: map_at_1000
value: 52.966
- type: map_at_3
value: 49.729
- type: map_at_5
value: 51.27
- type: mrr_at_1
value: 43.624
- type: mrr_at_10
value: 55.22899999999999
- type: mrr_at_100
value: 55.823
- type: mrr_at_1000
value: 55.85
- type: mrr_at_3
value: 52.739999999999995
- type: mrr_at_5
value: 54.251000000000005
- type: ndcg_at_1
value: 43.624
- type: ndcg_at_10
value: 58.23500000000001
- type: ndcg_at_100
value: 61.315
- type: ndcg_at_1000
value: 62.20099999999999
- type: ndcg_at_3
value: 53.22
- type: ndcg_at_5
value: 55.88999999999999
- type: precision_at_1
value: 43.624
- type: precision_at_10
value: 8.068999999999999
- type: precision_at_100
value: 0.975
- type: precision_at_1000
value: 0.107
- type: precision_at_3
value: 21.752
- type: precision_at_5
value: 14.515
- type: recall_at_1
value: 40.858
- type: recall_at_10
value: 73.744
- type: recall_at_100
value: 87.667
- type: recall_at_1000
value: 94.15599999999999
- type: recall_at_3
value: 60.287
- type: recall_at_5
value: 66.703
- task:
type: Retrieval
dataset:
type: fiqa
name: MTEB FiQA2018
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 17.864
- type: map_at_10
value: 28.592000000000002
- type: map_at_100
value: 30.165
- type: map_at_1000
value: 30.364
- type: map_at_3
value: 24.586
- type: map_at_5
value: 26.717000000000002
- type: mrr_at_1
value: 35.031
- type: mrr_at_10
value: 43.876
- type: mrr_at_100
value: 44.683
- type: mrr_at_1000
value: 44.736
- type: mrr_at_3
value: 40.998000000000005
- type: mrr_at_5
value: 42.595
- type: ndcg_at_1
value: 35.031
- type: ndcg_at_10
value: 36.368
- type: ndcg_at_100
value: 42.472
- type: ndcg_at_1000
value: 45.973000000000006
- type: ndcg_at_3
value: 31.915
- type: ndcg_at_5
value: 33.394
- type: precision_at_1
value: 35.031
- type: precision_at_10
value: 10.139
- type: precision_at_100
value: 1.6420000000000001
- type: precision_at_1000
value: 0.22699999999999998
- type: precision_at_3
value: 21.142
- type: precision_at_5
value: 15.772
- type: recall_at_1
value: 17.864
- type: recall_at_10
value: 43.991
- type: recall_at_100
value: 66.796
- type: recall_at_1000
value: 87.64
- type: recall_at_3
value: 28.915999999999997
- type: recall_at_5
value: 35.185
- task:
type: Retrieval
dataset:
type: hotpotqa
name: MTEB HotpotQA
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 36.556
- type: map_at_10
value: 53.056000000000004
- type: map_at_100
value: 53.909
- type: map_at_1000
value: 53.98
- type: map_at_3
value: 49.982
- type: map_at_5
value: 51.9
- type: mrr_at_1
value: 73.113
- type: mrr_at_10
value: 79.381
- type: mrr_at_100
value: 79.60300000000001
- type: mrr_at_1000
value: 79.617
- type: mrr_at_3
value: 78.298
- type: mrr_at_5
value: 78.995
- type: ndcg_at_1
value: 73.113
- type: ndcg_at_10
value: 62.21
- type: ndcg_at_100
value: 65.242
- type: ndcg_at_1000
value: 66.667
- type: ndcg_at_3
value: 57.717
- type: ndcg_at_5
value: 60.224
- type: precision_at_1
value: 73.113
- type: precision_at_10
value: 12.842999999999998
- type: precision_at_100
value: 1.522
- type: precision_at_1000
value: 0.17099999999999999
- type: precision_at_3
value: 36.178
- type: precision_at_5
value: 23.695
- type: recall_at_1
value: 36.556
- type: recall_at_10
value: 64.213
- type: recall_at_100
value: 76.077
- type: recall_at_1000
value: 85.53699999999999
- type: recall_at_3
value: 54.266999999999996
- type: recall_at_5
value: 59.236999999999995
- task:
type: Classification
dataset:
type: mteb/imdb
name: MTEB ImdbClassification
config: default
split: test
revision: 3d86128a09e091d6018b6d26cad27f2739fc2db7
metrics:
- type: accuracy
value: 75.958
- type: ap
value: 69.82869527654348
- type: f1
value: 75.89120903005633
- task:
type: Retrieval
dataset:
type: msmarco
name: MTEB MSMARCO
config: default
split: dev
revision: None
metrics:
- type: map_at_1
value: 23.608
- type: map_at_10
value: 36.144
- type: map_at_100
value: 37.244
- type: map_at_1000
value: 37.291999999999994
- type: map_at_3
value: 32.287
- type: map_at_5
value: 34.473
- type: mrr_at_1
value: 24.226
- type: mrr_at_10
value: 36.711
- type: mrr_at_100
value: 37.758
- type: mrr_at_1000
value: 37.8
- type: mrr_at_3
value: 32.92
- type: mrr_at_5
value: 35.104
- type: ndcg_at_1
value: 24.269
- type: ndcg_at_10
value: 43.138
- type: ndcg_at_100
value: 48.421
- type: ndcg_at_1000
value: 49.592000000000006
- type: ndcg_at_3
value: 35.269
- type: ndcg_at_5
value: 39.175
- type: precision_at_1
value: 24.269
- type: precision_at_10
value: 6.755999999999999
- type: precision_at_100
value: 0.941
- type: precision_at_1000
value: 0.104
- type: precision_at_3
value: 14.938
- type: precision_at_5
value: 10.934000000000001
- type: recall_at_1
value: 23.608
- type: recall_at_10
value: 64.679
- type: recall_at_100
value: 89.027
- type: recall_at_1000
value: 97.91
- type: recall_at_3
value: 43.25
- type: recall_at_5
value: 52.617000000000004
- task:
type: Classification
dataset:
type: mteb/mtop_domain
name: MTEB MTOPDomainClassification (en)
config: en
split: test
revision: d80d48c1eb48d3562165c59d59d0034df9fff0bf
metrics:
- type: accuracy
value: 93.21477428180576
- type: f1
value: 92.92502305092152
- task:
type: Classification
dataset:
type: mteb/mtop_intent
name: MTEB MTOPIntentClassification (en)
config: en
split: test
revision: ae001d0e6b1228650b7bd1c2c65fb50ad11a8aba
metrics:
- type: accuracy
value: 74.76744186046511
- type: f1
value: 59.19855520057899
- task:
type: Classification
dataset:
type: mteb/amazon_massive_intent
name: MTEB MassiveIntentClassification (en)
config: en
split: test
revision: 31efe3c427b0bae9c22cbb560b8f15491cc6bed7
metrics:
- type: accuracy
value: 72.24613315400134
- type: f1
value: 70.19950395651232
- task:
type: Classification
dataset:
type: mteb/amazon_massive_scenario
name: MTEB MassiveScenarioClassification (en)
config: en
split: test
revision: 7d571f92784cd94a019292a1f45445077d0ef634
metrics:
- type: accuracy
value: 76.75857431069268
- type: f1
value: 76.5433450230191
- task:
type: Clustering
dataset:
type: mteb/medrxiv-clustering-p2p
name: MTEB MedrxivClusteringP2P
config: default
split: test
revision: e7a26af6f3ae46b30dde8737f02c07b1505bcc73
metrics:
- type: v_measure
value: 31.525463791623604
- task:
type: Clustering
dataset:
type: mteb/medrxiv-clustering-s2s
name: MTEB MedrxivClusteringS2S
config: default
split: test
revision: 35191c8c0dca72d8ff3efcd72aa802307d469663
metrics:
- type: v_measure
value: 28.28695907385136
- task:
type: Reranking
dataset:
type: mteb/mind_small
name: MTEB MindSmallReranking
config: default
split: test
revision: 3bdac13927fdc888b903db93b2ffdbd90b295a69
metrics:
- type: map
value: 30.068174046665224
- type: mrr
value: 30.827586642840803
- task:
type: Retrieval
dataset:
type: nfcorpus
name: MTEB NFCorpus
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 6.322
- type: map_at_10
value: 13.919999999999998
- type: map_at_100
value: 17.416
- type: map_at_1000
value: 18.836
- type: map_at_3
value: 10.111
- type: map_at_5
value: 11.991999999999999
- type: mrr_at_1
value: 48.297000000000004
- type: mrr_at_10
value: 57.114
- type: mrr_at_100
value: 57.713
- type: mrr_at_1000
value: 57.751
- type: mrr_at_3
value: 55.108000000000004
- type: mrr_at_5
value: 56.533
- type: ndcg_at_1
value: 46.44
- type: ndcg_at_10
value: 36.589
- type: ndcg_at_100
value: 33.202
- type: ndcg_at_1000
value: 41.668
- type: ndcg_at_3
value: 41.302
- type: ndcg_at_5
value: 39.829
- type: precision_at_1
value: 47.988
- type: precision_at_10
value: 27.059
- type: precision_at_100
value: 8.235000000000001
- type: precision_at_1000
value: 2.091
- type: precision_at_3
value: 38.184000000000005
- type: precision_at_5
value: 34.365
- type: recall_at_1
value: 6.322
- type: recall_at_10
value: 18.288
- type: recall_at_100
value: 32.580999999999996
- type: recall_at_1000
value: 63.605999999999995
- type: recall_at_3
value: 11.266
- type: recall_at_5
value: 14.69
- task:
type: Retrieval
dataset:
type: nq
name: MTEB NQ
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 36.586999999999996
- type: map_at_10
value: 52.464
- type: map_at_100
value: 53.384
- type: map_at_1000
value: 53.405
- type: map_at_3
value: 48.408
- type: map_at_5
value: 50.788999999999994
- type: mrr_at_1
value: 40.904
- type: mrr_at_10
value: 54.974000000000004
- type: mrr_at_100
value: 55.60699999999999
- type: mrr_at_1000
value: 55.623
- type: mrr_at_3
value: 51.73799999999999
- type: mrr_at_5
value: 53.638
- type: ndcg_at_1
value: 40.904
- type: ndcg_at_10
value: 59.965999999999994
- type: ndcg_at_100
value: 63.613
- type: ndcg_at_1000
value: 64.064
- type: ndcg_at_3
value: 52.486
- type: ndcg_at_5
value: 56.377
- type: precision_at_1
value: 40.904
- type: precision_at_10
value: 9.551
- type: precision_at_100
value: 1.162
- type: precision_at_1000
value: 0.12
- type: precision_at_3
value: 23.552
- type: precision_at_5
value: 16.436999999999998
- type: recall_at_1
value: 36.586999999999996
- type: recall_at_10
value: 80.094
- type: recall_at_100
value: 95.515
- type: recall_at_1000
value: 98.803
- type: recall_at_3
value: 60.907
- type: recall_at_5
value: 69.817
- task:
type: Retrieval
dataset:
type: quora
name: MTEB QuoraRetrieval
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 70.422
- type: map_at_10
value: 84.113
- type: map_at_100
value: 84.744
- type: map_at_1000
value: 84.762
- type: map_at_3
value: 81.171
- type: map_at_5
value: 83.039
- type: mrr_at_1
value: 81.12
- type: mrr_at_10
value: 87.277
- type: mrr_at_100
value: 87.384
- type: mrr_at_1000
value: 87.385
- type: mrr_at_3
value: 86.315
- type: mrr_at_5
value: 86.981
- type: ndcg_at_1
value: 81.12
- type: ndcg_at_10
value: 87.92
- type: ndcg_at_100
value: 89.178
- type: ndcg_at_1000
value: 89.29899999999999
- type: ndcg_at_3
value: 85.076
- type: ndcg_at_5
value: 86.67099999999999
- type: precision_at_1
value: 81.12
- type: precision_at_10
value: 13.325999999999999
- type: precision_at_100
value: 1.524
- type: precision_at_1000
value: 0.157
- type: precision_at_3
value: 37.16
- type: precision_at_5
value: 24.456
- type: recall_at_1
value: 70.422
- type: recall_at_10
value: 95.00800000000001
- type: recall_at_100
value: 99.38
- type: recall_at_1000
value: 99.94800000000001
- type: recall_at_3
value: 86.809
- type: recall_at_5
value: 91.334
- task:
type: Clustering
dataset:
type: mteb/reddit-clustering
name: MTEB RedditClustering
config: default
split: test
revision: 24640382cdbf8abc73003fb0fa6d111a705499eb
metrics:
- type: v_measure
value: 48.18491891699636
- task:
type: Clustering
dataset:
type: mteb/reddit-clustering-p2p
name: MTEB RedditClusteringP2P
config: default
split: test
revision: 282350215ef01743dc01b456c7f5241fa8937f16
metrics:
- type: v_measure
value: 62.190639679711914
- task:
type: Retrieval
dataset:
type: scidocs
name: MTEB SCIDOCS
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 4.478
- type: map_at_10
value: 11.268
- type: map_at_100
value: 13.129
- type: map_at_1000
value: 13.41
- type: map_at_3
value: 8.103
- type: map_at_5
value: 9.609
- type: mrr_at_1
value: 22
- type: mrr_at_10
value: 32.248
- type: mrr_at_100
value: 33.355000000000004
- type: mrr_at_1000
value: 33.42
- type: mrr_at_3
value: 29.15
- type: mrr_at_5
value: 30.785
- type: ndcg_at_1
value: 22
- type: ndcg_at_10
value: 18.990000000000002
- type: ndcg_at_100
value: 26.302999999999997
- type: ndcg_at_1000
value: 31.537
- type: ndcg_at_3
value: 18.034
- type: ndcg_at_5
value: 15.655
- type: precision_at_1
value: 22
- type: precision_at_10
value: 9.91
- type: precision_at_100
value: 2.0420000000000003
- type: precision_at_1000
value: 0.33
- type: precision_at_3
value: 16.933
- type: precision_at_5
value: 13.719999999999999
- type: recall_at_1
value: 4.478
- type: recall_at_10
value: 20.087
- type: recall_at_100
value: 41.457
- type: recall_at_1000
value: 67.10199999999999
- type: recall_at_3
value: 10.313
- type: recall_at_5
value: 13.927999999999999
- task:
type: STS
dataset:
type: mteb/sickr-sts
name: MTEB SICK-R
config: default
split: test
revision: a6ea5a8cab320b040a23452cc28066d9beae2cee
metrics:
- type: cos_sim_pearson
value: 84.27341574565806
- type: cos_sim_spearman
value: 79.66419880841734
- type: euclidean_pearson
value: 81.32473321838208
- type: euclidean_spearman
value: 79.29828832085133
- type: manhattan_pearson
value: 81.25554065883132
- type: manhattan_spearman
value: 79.23275543279853
- task:
type: STS
dataset:
type: mteb/sts12-sts
name: MTEB STS12
config: default
split: test
revision: a0d554a64d88156834ff5ae9920b964011b16384
metrics:
- type: cos_sim_pearson
value: 83.40468875905418
- type: cos_sim_spearman
value: 74.2189990321174
- type: euclidean_pearson
value: 80.74376966290956
- type: euclidean_spearman
value: 74.97663839079335
- type: manhattan_pearson
value: 80.69779331646207
- type: manhattan_spearman
value: 75.00225252917613
- task:
type: STS
dataset:
type: mteb/sts13-sts
name: MTEB STS13
config: default
split: test
revision: 7e90230a92c190f1bf69ae9002b8cea547a64cca
metrics:
- type: cos_sim_pearson
value: 82.5745290053095
- type: cos_sim_spearman
value: 83.31401180333397
- type: euclidean_pearson
value: 82.96500607325534
- type: euclidean_spearman
value: 83.8534967935793
- type: manhattan_pearson
value: 82.83112050632508
- type: manhattan_spearman
value: 83.70877296557838
- task:
type: STS
dataset:
type: mteb/sts14-sts
name: MTEB STS14
config: default
split: test
revision: 6031580fec1f6af667f0bd2da0a551cf4f0b2375
metrics:
- type: cos_sim_pearson
value: 80.67833656607704
- type: cos_sim_spearman
value: 78.52252410630707
- type: euclidean_pearson
value: 80.071189514343
- type: euclidean_spearman
value: 78.95143545742796
- type: manhattan_pearson
value: 80.0128926165121
- type: manhattan_spearman
value: 78.91236678732628
- task:
type: STS
dataset:
type: mteb/sts15-sts
name: MTEB STS15
config: default
split: test
revision: ae752c7c21bf194d8b67fd573edf7ae58183cbe3
metrics:
- type: cos_sim_pearson
value: 87.48437639980746
- type: cos_sim_spearman
value: 88.34876527774259
- type: euclidean_pearson
value: 87.64898081823888
- type: euclidean_spearman
value: 88.58937180804213
- type: manhattan_pearson
value: 87.5942417815288
- type: manhattan_spearman
value: 88.53013922267687
- task:
type: STS
dataset:
type: mteb/sts16-sts
name: MTEB STS16
config: default
split: test
revision: 4d8694f8f0e0100860b497b999b3dbed754a0513
metrics:
- type: cos_sim_pearson
value: 82.69189187164781
- type: cos_sim_spearman
value: 84.15327883572112
- type: euclidean_pearson
value: 83.64202266685898
- type: euclidean_spearman
value: 84.6219602318862
- type: manhattan_pearson
value: 83.53256698709998
- type: manhattan_spearman
value: 84.49260712904946
- task:
type: STS
dataset:
type: mteb/sts17-crosslingual-sts
name: MTEB STS17 (en-en)
config: en-en
split: test
revision: af5e6fb845001ecf41f4c1e033ce921939a2a68d
metrics:
- type: cos_sim_pearson
value: 87.09508017611589
- type: cos_sim_spearman
value: 87.23010990417097
- type: euclidean_pearson
value: 87.62545569077133
- type: euclidean_spearman
value: 86.71152051711714
- type: manhattan_pearson
value: 87.5057154278377
- type: manhattan_spearman
value: 86.60611898281267
- task:
type: STS
dataset:
type: mteb/sts22-crosslingual-sts
name: MTEB STS22 (en)
config: en
split: test
revision: 6d1ba47164174a496b7fa5d3569dae26a6813b80
metrics:
- type: cos_sim_pearson
value: 61.72129893941176
- type: cos_sim_spearman
value: 62.87871412069194
- type: euclidean_pearson
value: 63.21077648290454
- type: euclidean_spearman
value: 63.03263080805978
- type: manhattan_pearson
value: 63.20740860135976
- type: manhattan_spearman
value: 62.89930471802817
- task:
type: STS
dataset:
type: mteb/stsbenchmark-sts
name: MTEB STSBenchmark
config: default
split: test
revision: b0fddb56ed78048fa8b90373c8a3cfc37b684831
metrics:
- type: cos_sim_pearson
value: 85.039118236799
- type: cos_sim_spearman
value: 86.18102563389962
- type: euclidean_pearson
value: 85.62977041471879
- type: euclidean_spearman
value: 86.02478990544347
- type: manhattan_pearson
value: 85.60786740521806
- type: manhattan_spearman
value: 85.99546210442547
- task:
type: Reranking
dataset:
type: mteb/scidocs-reranking
name: MTEB SciDocsRR
config: default
split: test
revision: d3c5e1fc0b855ab6097bf1cda04dd73947d7caab
metrics:
- type: map
value: 82.89875069737266
- type: mrr
value: 95.42621322033087
- task:
type: Retrieval
dataset:
type: scifact
name: MTEB SciFact
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 58.660999999999994
- type: map_at_10
value: 68.738
- type: map_at_100
value: 69.33200000000001
- type: map_at_1000
value: 69.352
- type: map_at_3
value: 66.502
- type: map_at_5
value: 67.686
- type: mrr_at_1
value: 61.667
- type: mrr_at_10
value: 70.003
- type: mrr_at_100
value: 70.441
- type: mrr_at_1000
value: 70.46
- type: mrr_at_3
value: 68.278
- type: mrr_at_5
value: 69.194
- type: ndcg_at_1
value: 61.667
- type: ndcg_at_10
value: 73.083
- type: ndcg_at_100
value: 75.56
- type: ndcg_at_1000
value: 76.01400000000001
- type: ndcg_at_3
value: 69.28699999999999
- type: ndcg_at_5
value: 70.85000000000001
- type: precision_at_1
value: 61.667
- type: precision_at_10
value: 9.6
- type: precision_at_100
value: 1.087
- type: precision_at_1000
value: 0.11199999999999999
- type: precision_at_3
value: 27.111
- type: precision_at_5
value: 17.467
- type: recall_at_1
value: 58.660999999999994
- type: recall_at_10
value: 85.02199999999999
- type: recall_at_100
value: 95.933
- type: recall_at_1000
value: 99.333
- type: recall_at_3
value: 74.506
- type: recall_at_5
value: 78.583
- task:
type: PairClassification
dataset:
type: mteb/sprintduplicatequestions-pairclassification
name: MTEB SprintDuplicateQuestions
config: default
split: test
revision: d66bd1f72af766a5cc4b0ca5e00c162f89e8cc46
metrics:
- type: cos_sim_accuracy
value: 99.8029702970297
- type: cos_sim_ap
value: 94.87673936635738
- type: cos_sim_f1
value: 90.00502260170768
- type: cos_sim_precision
value: 90.41372351160445
- type: cos_sim_recall
value: 89.60000000000001
- type: dot_accuracy
value: 99.57524752475247
- type: dot_ap
value: 84.81717934496321
- type: dot_f1
value: 78.23026646556059
- type: dot_precision
value: 78.66531850353893
- type: dot_recall
value: 77.8
- type: euclidean_accuracy
value: 99.8029702970297
- type: euclidean_ap
value: 94.74658253135284
- type: euclidean_f1
value: 90.08470353761834
- type: euclidean_precision
value: 89.77159880834161
- type: euclidean_recall
value: 90.4
- type: manhattan_accuracy
value: 99.8
- type: manhattan_ap
value: 94.69224030742787
- type: manhattan_f1
value: 89.9502487562189
- type: manhattan_precision
value: 89.50495049504951
- type: manhattan_recall
value: 90.4
- type: max_accuracy
value: 99.8029702970297
- type: max_ap
value: 94.87673936635738
- type: max_f1
value: 90.08470353761834
- task:
type: Clustering
dataset:
type: mteb/stackexchange-clustering
name: MTEB StackExchangeClustering
config: default
split: test
revision: 6cbc1f7b2bc0622f2e39d2c77fa502909748c259
metrics:
- type: v_measure
value: 63.906039623153035
- task:
type: Clustering
dataset:
type: mteb/stackexchange-clustering-p2p
name: MTEB StackExchangeClusteringP2P
config: default
split: test
revision: 815ca46b2622cec33ccafc3735d572c266efdb44
metrics:
- type: v_measure
value: 32.56053830923281
- task:
type: Reranking
dataset:
type: mteb/stackoverflowdupquestions-reranking
name: MTEB StackOverflowDupQuestions
config: default
split: test
revision: e185fbe320c72810689fc5848eb6114e1ef5ec69
metrics:
- type: map
value: 50.15326538775145
- type: mrr
value: 50.99279295051355
- task:
type: Summarization
dataset:
type: mteb/summeval
name: MTEB SummEval
config: default
split: test
revision: cda12ad7615edc362dbf25a00fdd61d3b1eaf93c
metrics:
- type: cos_sim_pearson
value: 31.44030762047337
- type: cos_sim_spearman
value: 31.00910300264562
- type: dot_pearson
value: 26.88257194766013
- type: dot_spearman
value: 27.646202679013577
- task:
type: Retrieval
dataset:
type: trec-covid
name: MTEB TRECCOVID
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 0.247
- type: map_at_10
value: 1.9429999999999998
- type: map_at_100
value: 10.82
- type: map_at_1000
value: 25.972
- type: map_at_3
value: 0.653
- type: map_at_5
value: 1.057
- type: mrr_at_1
value: 94
- type: mrr_at_10
value: 96.333
- type: mrr_at_100
value: 96.333
- type: mrr_at_1000
value: 96.333
- type: mrr_at_3
value: 96.333
- type: mrr_at_5
value: 96.333
- type: ndcg_at_1
value: 89
- type: ndcg_at_10
value: 79.63799999999999
- type: ndcg_at_100
value: 57.961
- type: ndcg_at_1000
value: 50.733
- type: ndcg_at_3
value: 84.224
- type: ndcg_at_5
value: 82.528
- type: precision_at_1
value: 94
- type: precision_at_10
value: 84.2
- type: precision_at_100
value: 59.36
- type: precision_at_1000
value: 22.738
- type: precision_at_3
value: 88
- type: precision_at_5
value: 86.8
- type: recall_at_1
value: 0.247
- type: recall_at_10
value: 2.131
- type: recall_at_100
value: 14.035
- type: recall_at_1000
value: 47.457
- type: recall_at_3
value: 0.6779999999999999
- type: recall_at_5
value: 1.124
- task:
type: Retrieval
dataset:
type: webis-touche2020
name: MTEB Touche2020
config: default
split: test
revision: None
metrics:
- type: map_at_1
value: 2.603
- type: map_at_10
value: 11.667
- type: map_at_100
value: 16.474
- type: map_at_1000
value: 18.074
- type: map_at_3
value: 6.03
- type: map_at_5
value: 8.067
- type: mrr_at_1
value: 34.694
- type: mrr_at_10
value: 51.063
- type: mrr_at_100
value: 51.908
- type: mrr_at_1000
value: 51.908
- type: mrr_at_3
value: 47.959
- type: mrr_at_5
value: 49.694
- type: ndcg_at_1
value: 32.653
- type: ndcg_at_10
value: 28.305000000000003
- type: ndcg_at_100
value: 35.311
- type: ndcg_at_1000
value: 47.644999999999996
- type: ndcg_at_3
value: 32.187
- type: ndcg_at_5
value: 29.134999999999998
- type: precision_at_1
value: 34.694
- type: precision_at_10
value: 26.122
- type: precision_at_100
value: 6.755
- type: precision_at_1000
value: 1.467
- type: precision_at_3
value: 34.694
- type: precision_at_5
value: 30.203999999999997
- type: recall_at_1
value: 2.603
- type: recall_at_10
value: 18.716
- type: recall_at_100
value: 42.512
- type: recall_at_1000
value: 79.32000000000001
- type: recall_at_3
value: 7.59
- type: recall_at_5
value: 10.949
- task:
type: Classification
dataset:
type: mteb/toxic_conversations_50k
name: MTEB ToxicConversationsClassification
config: default
split: test
revision: d7c0de2777da35d6aae2200a62c6e0e5af397c4c
metrics:
- type: accuracy
value: 74.117
- type: ap
value: 15.89357321699319
- type: f1
value: 57.14385866369257
- task:
type: Classification
dataset:
type: mteb/tweet_sentiment_extraction
name: MTEB TweetSentimentExtractionClassification
config: default
split: test
revision: d604517c81ca91fe16a244d1248fc021f9ecee7a
metrics:
- type: accuracy
value: 61.38370118845502
- type: f1
value: 61.67038693866553
- task:
type: Clustering
dataset:
type: mteb/twentynewsgroups-clustering
name: MTEB TwentyNewsgroupsClustering
config: default
split: test
revision: 6125ec4e24fa026cec8a478383ee943acfbd5449
metrics:
- type: v_measure
value: 42.57754941537969
- task:
type: PairClassification
dataset:
type: mteb/twittersemeval2015-pairclassification
name: MTEB TwitterSemEval2015
config: default
split: test
revision: 70970daeab8776df92f5ea462b6173c0b46fd2d1
metrics:
- type: cos_sim_accuracy
value: 86.1775049174465
- type: cos_sim_ap
value: 74.3994879581554
- type: cos_sim_f1
value: 69.32903671308551
- type: cos_sim_precision
value: 61.48193508879363
- type: cos_sim_recall
value: 79.47229551451187
- type: dot_accuracy
value: 81.65345413363534
- type: dot_ap
value: 59.690898346685096
- type: dot_f1
value: 57.27622826467499
- type: dot_precision
value: 51.34965473948525
- type: dot_recall
value: 64.74934036939314
- type: euclidean_accuracy
value: 86.04637301066937
- type: euclidean_ap
value: 74.33009001775268
- type: euclidean_f1
value: 69.2458374142997
- type: euclidean_precision
value: 64.59570580173595
- type: euclidean_recall
value: 74.6174142480211
- type: manhattan_accuracy
value: 86.11193896405793
- type: manhattan_ap
value: 74.2964140130421
- type: manhattan_f1
value: 69.11601528788066
- type: manhattan_precision
value: 64.86924323073363
- type: manhattan_recall
value: 73.95778364116094
- type: max_accuracy
value: 86.1775049174465
- type: max_ap
value: 74.3994879581554
- type: max_f1
value: 69.32903671308551
- task:
type: PairClassification
dataset:
type: mteb/twitterurlcorpus-pairclassification
name: MTEB TwitterURLCorpus
config: default
split: test
revision: 8b6510b0b1fa4e4c4f879467980e9be563ec1cdf
metrics:
- type: cos_sim_accuracy
value: 89.01501921061823
- type: cos_sim_ap
value: 85.97819287477351
- type: cos_sim_f1
value: 78.33882858518875
- type: cos_sim_precision
value: 75.49446626204926
- type: cos_sim_recall
value: 81.40591315060055
- type: dot_accuracy
value: 86.47494857763806
- type: dot_ap
value: 78.77420360340282
- type: dot_f1
value: 73.06433247936238
- type: dot_precision
value: 67.92140777983595
- type: dot_recall
value: 79.04989220819218
- type: euclidean_accuracy
value: 88.7297706368611
- type: euclidean_ap
value: 85.61550568529317
- type: euclidean_f1
value: 77.84805525263539
- type: euclidean_precision
value: 73.73639994491117
- type: euclidean_recall
value: 82.44533415460425
- type: manhattan_accuracy
value: 88.75111576823068
- type: manhattan_ap
value: 85.58701671476263
- type: manhattan_f1
value: 77.70169909067856
- type: manhattan_precision
value: 73.37666780704755
- type: manhattan_recall
value: 82.5685247921158
- type: max_accuracy
value: 89.01501921061823
- type: max_ap
value: 85.97819287477351
- type: max_f1
value: 78.33882858518875
language:
- en
license: mit
---
## E5-base
**News (May 2023): please switch to [e5-base-v2](https://huggingface.co/intfloat/e5-base-v2), which has better performance and same method of usage.**
[Text Embeddings by Weakly-Supervised Contrastive Pre-training](https://arxiv.org/pdf/2212.03533.pdf).
Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, Furu Wei, arXiv 2022
This model has 12 layers and the embedding size is 768.
## Usage
Below is an example to encode queries and passages from the MS-MARCO passage ranking dataset.
```python
import torch.nn.functional as F
from torch import Tensor
from transformers import AutoTokenizer, AutoModel
def average_pool(last_hidden_states: Tensor,
attention_mask: Tensor) -> Tensor:
last_hidden = last_hidden_states.masked_fill(~attention_mask[..., None].bool(), 0.0)
return last_hidden.sum(dim=1) / attention_mask.sum(dim=1)[..., None]
# Each input text should start with "query: " or "passage: ".
# For tasks other than retrieval, you can simply use the "query: " prefix.
input_texts = ['query: how much protein should a female eat',
'query: summit define',
"passage: As a general guideline, the CDC's average requirement of protein for women ages 19 to 70 is 46 grams per day. But, as you can see from this chart, you'll need to increase that if you're expecting or training for a marathon. Check out the chart below to see how much protein you should be eating each day.",
"passage: Definition of summit for English Language Learners. : 1 the highest point of a mountain : the top of a mountain. : 2 the highest level. : 3 a meeting or series of meetings between the leaders of two or more governments."]
tokenizer = AutoTokenizer.from_pretrained('intfloat/e5-base')
model = AutoModel.from_pretrained('intfloat/e5-base')
# Tokenize the input texts
batch_dict = tokenizer(input_texts, max_length=512, padding=True, truncation=True, return_tensors='pt')
outputs = model(**batch_dict)
embeddings = average_pool(outputs.last_hidden_state, batch_dict['attention_mask'])
# normalize embeddings
embeddings = F.normalize(embeddings, p=2, dim=1)
scores = (embeddings[:2] @ embeddings[2:].T) * 100
print(scores.tolist())
```
## Training Details
Please refer to our paper at [https://arxiv.org/pdf/2212.03533.pdf](https://arxiv.org/pdf/2212.03533.pdf).
## Benchmark Evaluation
Check out [unilm/e5](https://github.com/microsoft/unilm/tree/master/e5) to reproduce evaluation results
on the [BEIR](https://arxiv.org/abs/2104.08663) and [MTEB benchmark](https://arxiv.org/abs/2210.07316).
## Support for Sentence Transformers
Below is an example for usage with sentence_transformers.
```python
from sentence_transformers import SentenceTransformer
model = SentenceTransformer('intfloat/e5-base')
input_texts = [
'query: how much protein should a female eat',
'query: summit define',
"passage: As a general guideline, the CDC's average requirement of protein for women ages 19 to 70 is 46 grams per day. But, as you can see from this chart, you'll need to increase that if you're expecting or training for a marathon. Check out the chart below to see how much protein you should be eating each day.",
"passage: Definition of summit for English Language Learners. : 1 the highest point of a mountain : the top of a mountain. : 2 the highest level. : 3 a meeting or series of meetings between the leaders of two or more governments."
]
embeddings = model.encode(input_texts, normalize_embeddings=True)
```
Package requirements
`pip install sentence_transformers~=2.2.2`
Contributors: [michaelfeil](https://huggingface.co/michaelfeil)
## FAQ
**1. Do I need to add the prefix "query: " and "passage: " to input texts?**
Yes, this is how the model is trained, otherwise you will see a performance degradation.
Here are some rules of thumb:
- Use "query: " and "passage: " correspondingly for asymmetric tasks such as passage retrieval in open QA, ad-hoc information retrieval.
- Use "query: " prefix for symmetric tasks such as semantic similarity, paraphrase retrieval.
- Use "query: " prefix if you want to use embeddings as features, such as linear probing classification, clustering.
**2. Why are my reproduced results slightly different from reported in the model card?**
Different versions of `transformers` and `pytorch` could cause negligible but non-zero performance differences.
**3. Why does the cosine similarity scores distribute around 0.7 to 1.0?**
This is a known and expected behavior as we use a low temperature 0.01 for InfoNCE contrastive loss.
For text embedding tasks like text retrieval or semantic similarity,
what matters is the relative order of the scores instead of the absolute values,
so this should not be an issue.
## Citation
If you find our paper or models helpful, please consider cite as follows:
```
@article{wang2022text,
title={Text Embeddings by Weakly-Supervised Contrastive Pre-training},
author={Wang, Liang and Yang, Nan and Huang, Xiaolong and Jiao, Binxing and Yang, Linjun and Jiang, Daxin and Majumder, Rangan and Wei, Furu},
journal={arXiv preprint arXiv:2212.03533},
year={2022}
}
```
## Limitations
This model only works for English texts. Long texts will be truncated to at most 512 tokens.
| 67,552 | [
[
-0.0092926025390625,
-0.052825927734375,
0.01271820068359375,
0.0181884765625,
-0.0196685791015625,
-0.03460693359375,
0.0015659332275390625,
-0.032806396484375,
0.004161834716796875,
0.0232086181640625,
-0.037689208984375,
-0.047119140625,
-0.07501220703125,
0.0196380615234375,
-0.0268707275390625,
0.07049560546875,
0.0011615753173828125,
0.0074615478515625,
-0.0289459228515625,
-0.00418853759765625,
-0.01898193359375,
-0.043701171875,
-0.0250701904296875,
-0.025390625,
0.0225067138671875,
0.017425537109375,
0.042572021484375,
0.045166015625,
0.053466796875,
0.0256195068359375,
-0.01033782958984375,
0.011810302734375,
-0.0413818359375,
-0.0116729736328125,
-0.00027561187744140625,
-0.0408935546875,
-0.03472900390625,
0.015472412109375,
0.03424072265625,
0.06475830078125,
0.01220703125,
0.0212860107421875,
0.0275726318359375,
0.043853759765625,
-0.04644775390625,
0.014312744140625,
-0.031524658203125,
0.0104827880859375,
0.00921630859375,
-0.0027942657470703125,
-0.0294952392578125,
0.0127105712890625,
0.0291290283203125,
-0.044525146484375,
0.02288818359375,
0.010345458984375,
0.09674072265625,
0.024078369140625,
-0.035125732421875,
-0.01395416259765625,
-0.00913238525390625,
0.0753173828125,
-0.052154541015625,
0.037506103515625,
0.05126953125,
-0.0198516845703125,
-0.0072479248046875,
-0.071044921875,
-0.027130126953125,
-0.01447296142578125,
-0.017303466796875,
0.01346588134765625,
-0.0200958251953125,
-0.00403594970703125,
0.0313720703125,
0.03375244140625,
-0.0628662109375,
-0.008453369140625,
-0.0279998779296875,
-0.007183074951171875,
0.037872314453125,
0.00847625732421875,
0.022796630859375,
-0.0355224609375,
-0.0172882080078125,
-0.0210723876953125,
-0.042633056640625,
0.0036869049072265625,
0.01525115966796875,
0.0299530029296875,
-0.029998779296875,
0.041534423828125,
-0.023193359375,
0.046630859375,
0.0175933837890625,
0.0087738037109375,
0.053558349609375,
-0.0404052734375,
-0.02056884765625,
-0.0188140869140625,
0.0712890625,
0.042724609375,
0.01024627685546875,
-0.0092315673828125,
-0.00156402587890625,
-0.004093170166015625,
0.006237030029296875,
-0.08526611328125,
-0.041015625,
0.0158538818359375,
-0.051666259765625,
-0.01593017578125,
0.009521484375,
-0.04156494140625,
-0.00405120849609375,
-0.0196685791015625,
0.0643310546875,
-0.042816162109375,
0.004974365234375,
0.020355224609375,
-0.0181121826171875,
0.00936126708984375,
0.00994873046875,
-0.06597900390625,
0.0255889892578125,
0.00885009765625,
0.0670166015625,
-0.007236480712890625,
-0.0293731689453125,
-0.036376953125,
-0.006565093994140625,
0.00435638427734375,
0.037933349609375,
-0.0266571044921875,
-0.0200347900390625,
0.0006885528564453125,
0.0301513671875,
-0.035491943359375,
-0.038909912109375,
0.041015625,
-0.0228271484375,
0.03424072265625,
-0.0214691162109375,
-0.041015625,
-0.00421905517578125,
0.01873779296875,
-0.0318603515625,
0.0823974609375,
0.007640838623046875,
-0.07135009765625,
0.0113372802734375,
-0.03350830078125,
-0.029052734375,
-0.01361083984375,
-0.004314422607421875,
-0.03631591796875,
-0.01068115234375,
0.038055419921875,
0.028594970703125,
-0.0087738037109375,
0.0005116462707519531,
-0.003749847412109375,
-0.038848876953125,
0.01195526123046875,
-0.012451171875,
0.06707763671875,
0.008270263671875,
-0.035980224609375,
-0.01384735107421875,
-0.05548095703125,
0.00566864013671875,
0.00998687744140625,
-0.0355224609375,
-0.0092620849609375,
0.00858306884765625,
0.0008654594421386719,
0.0218963623046875,
0.0289459228515625,
-0.03558349609375,
0.0134735107421875,
-0.04119873046875,
0.052825927734375,
0.039703369140625,
0.006885528564453125,
0.037139892578125,
-0.031768798828125,
0.00859832763671875,
0.02801513671875,
0.005489349365234375,
-0.0009241104125976562,
-0.040069580078125,
-0.059906005859375,
-0.00730133056640625,
0.043365478515625,
0.041351318359375,
-0.033599853515625,
0.044097900390625,
-0.0258331298828125,
-0.0224151611328125,
-0.050689697265625,
0.006267547607421875,
0.01751708984375,
0.030517578125,
0.061981201171875,
-0.00952911376953125,
-0.054718017578125,
-0.07379150390625,
-0.024871826171875,
0.0091400146484375,
-0.0197906494140625,
0.018951416015625,
0.0655517578125,
-0.0252838134765625,
0.046630859375,
-0.049591064453125,
-0.033294677734375,
-0.0172576904296875,
0.00821685791015625,
0.0287933349609375,
0.05841064453125,
0.0299530029296875,
-0.06689453125,
-0.03533935546875,
-0.040771484375,
-0.06817626953125,
0.0022735595703125,
0.00907135009765625,
-0.0194091796875,
-0.00518798828125,
0.036529541015625,
-0.048828125,
0.02557373046875,
0.03680419921875,
-0.034423828125,
0.0188446044921875,
-0.0194244384765625,
0.01222991943359375,
-0.079833984375,
-0.0001436471939086914,
0.01287841796875,
-0.01526641845703125,
-0.0294342041015625,
0.01053619384765625,
0.0017337799072265625,
-0.00897216796875,
-0.037261962890625,
0.0211334228515625,
-0.044525146484375,
0.0181121826171875,
-0.006885528564453125,
0.0225830078125,
0.0242462158203125,
0.03955078125,
-0.0076141357421875,
0.042999267578125,
0.044158935546875,
-0.06597900390625,
-0.0006089210510253906,
0.05169677734375,
-0.0257568359375,
0.0234832763671875,
-0.06524658203125,
0.00873565673828125,
-0.002574920654296875,
0.01788330078125,
-0.0701904296875,
-0.011383056640625,
0.0241241455078125,
-0.05389404296875,
0.0239410400390625,
-0.0018815994262695312,
-0.03582763671875,
-0.02032470703125,
-0.039764404296875,
0.0176239013671875,
0.0416259765625,
-0.0266571044921875,
0.0372314453125,
0.0181121826171875,
0.002849578857421875,
-0.042449951171875,
-0.08056640625,
-0.00905609130859375,
-0.00144195556640625,
-0.0509033203125,
0.05712890625,
-0.01418304443359375,
0.0143280029296875,
0.005329132080078125,
-0.0147552490234375,
0.0160064697265625,
-0.0099639892578125,
0.020172119140625,
0.0028018951416015625,
-0.0004475116729736328,
0.0041961669921875,
-0.00965118408203125,
-0.004062652587890625,
0.0020599365234375,
-0.0207061767578125,
0.042999267578125,
-0.023284912109375,
0.0082550048828125,
-0.041656494140625,
0.03839111328125,
0.01256561279296875,
-0.0175018310546875,
0.0814208984375,
0.058837890625,
-0.0289306640625,
0.008026123046875,
-0.0223541259765625,
-0.02679443359375,
-0.035919189453125,
0.047882080078125,
-0.042633056640625,
-0.039154052734375,
0.0296630859375,
0.005710601806640625,
-0.00695037841796875,
0.06732177734375,
0.0254669189453125,
-0.0224761962890625,
0.097412109375,
0.05615234375,
0.01215362548828125,
0.033538818359375,
-0.052947998046875,
0.00362396240234375,
-0.07159423828125,
-0.0266571044921875,
-0.04559326171875,
-0.032989501953125,
-0.06298828125,
-0.0312347412109375,
0.02337646484375,
0.0180511474609375,
-0.035400390625,
0.0294189453125,
-0.043670654296875,
0.0060577392578125,
0.039642333984375,
0.041656494140625,
0.003910064697265625,
0.01361083984375,
-0.0137939453125,
-0.030059814453125,
-0.06964111328125,
-0.0284881591796875,
0.0771484375,
0.0212249755859375,
0.054229736328125,
-0.0021991729736328125,
0.04913330078125,
0.01074981689453125,
-0.007415771484375,
-0.0484619140625,
0.042327880859375,
-0.0289306640625,
-0.0238494873046875,
-0.00914764404296875,
-0.051727294921875,
-0.0775146484375,
0.032928466796875,
-0.0302734375,
-0.05157470703125,
0.01467132568359375,
-0.010345458984375,
-0.0172119140625,
0.006805419921875,
-0.0689697265625,
0.079833984375,
0.0026702880859375,
-0.025146484375,
0.00211334228515625,
-0.05230712890625,
-0.0184783935546875,
0.0307769775390625,
0.00899505615234375,
0.000667572021484375,
-0.00487518310546875,
0.07940673828125,
-0.023040771484375,
0.069580078125,
-0.00740814208984375,
0.032470703125,
0.006404876708984375,
-0.01264190673828125,
0.040130615234375,
-0.01432037353515625,
-0.007236480712890625,
0.0194091796875,
0.0030002593994140625,
-0.04400634765625,
-0.0251312255859375,
0.0584716796875,
-0.09161376953125,
-0.040252685546875,
-0.040985107421875,
-0.0362548828125,
0.0055999755859375,
0.0111236572265625,
0.05072021484375,
0.034332275390625,
0.00965118408203125,
0.04052734375,
0.0423583984375,
-0.025543212890625,
0.0243072509765625,
0.0225067138671875,
0.0091552734375,
-0.032440185546875,
0.053314208984375,
0.0301055908203125,
0.0141143798828125,
0.051971435546875,
0.016998291015625,
-0.0243072509765625,
-0.042449951171875,
-0.013153076171875,
0.0301513671875,
-0.054443359375,
-0.01392364501953125,
-0.084716796875,
-0.0227508544921875,
-0.046966552734375,
0.001262664794921875,
-0.0188446044921875,
-0.0318603515625,
-0.0299530029296875,
-0.006458282470703125,
0.01275634765625,
0.0265045166015625,
-0.004512786865234375,
0.021331787109375,
-0.04852294921875,
0.02447509765625,
0.0010738372802734375,
0.004222869873046875,
-0.013885498046875,
-0.071044921875,
-0.03082275390625,
0.00872039794921875,
-0.04559326171875,
-0.07037353515625,
0.031768798828125,
0.033233642578125,
0.047393798828125,
0.003322601318359375,
0.006519317626953125,
0.0484619140625,
-0.026702880859375,
0.07366943359375,
0.00832366943359375,
-0.067626953125,
0.05072021484375,
-0.005268096923828125,
0.050140380859375,
0.03948974609375,
0.0567626953125,
-0.02630615234375,
-0.0282135009765625,
-0.0584716796875,
-0.08502197265625,
0.048309326171875,
0.0322265625,
0.0157470703125,
-0.006015777587890625,
0.0227203369140625,
-0.0018968582153320312,
0.0188751220703125,
-0.08172607421875,
-0.030670166015625,
-0.030853271484375,
-0.02447509765625,
-0.00872039794921875,
-0.01554107666015625,
0.0022182464599609375,
-0.0439453125,
0.06427001953125,
-0.0013332366943359375,
0.05023193359375,
0.042633056640625,
-0.03839111328125,
0.0049285888671875,
0.0037937164306640625,
0.02056884765625,
0.0435791015625,
-0.03558349609375,
0.0219573974609375,
0.02685546875,
-0.050048828125,
-0.0120849609375,
0.016510009765625,
-0.0182647705078125,
0.005649566650390625,
0.035675048828125,
0.05487060546875,
0.024169921875,
-0.022979736328125,
0.04437255859375,
0.0019483566284179688,
-0.0269317626953125,
-0.0055084228515625,
-0.002162933349609375,
0.01285552978515625,
0.013092041015625,
0.03265380859375,
0.001277923583984375,
0.01323699951171875,
-0.044830322265625,
0.00925445556640625,
-0.005222320556640625,
-0.0301055908203125,
-0.0218658447265625,
0.053131103515625,
0.0183563232421875,
-0.00783538818359375,
0.07501220703125,
-0.00946807861328125,
-0.041351318359375,
0.036224365234375,
0.05364990234375,
0.0491943359375,
-0.0083465576171875,
0.01145172119140625,
0.06622314453125,
0.0291290283203125,
0.0023441314697265625,
0.0172576904296875,
0.01360321044921875,
-0.052032470703125,
-0.0183868408203125,
-0.06671142578125,
-0.004608154296875,
0.016876220703125,
-0.038055419921875,
0.01495361328125,
-0.00461578369140625,
-0.0225982666015625,
0.0009908676147460938,
0.0338134765625,
-0.07086181640625,
0.020599365234375,
-0.003070831298828125,
0.055999755859375,
-0.06732177734375,
0.0382080078125,
0.06060791015625,
-0.0634765625,
-0.053009033203125,
-0.000835418701171875,
-0.0242767333984375,
-0.04107666015625,
0.054168701171875,
0.03759765625,
0.007358551025390625,
0.00566864013671875,
-0.040283203125,
-0.0537109375,
0.08648681640625,
0.0139312744140625,
-0.03533935546875,
-0.023681640625,
0.0230712890625,
0.031585693359375,
-0.037750244140625,
0.040069580078125,
0.0262298583984375,
0.022186279296875,
-0.0079498291015625,
-0.050628662109375,
0.01922607421875,
-0.0291290283203125,
-0.00666046142578125,
-0.01303863525390625,
-0.050140380859375,
0.08734130859375,
-0.018463134765625,
-0.00514984130859375,
0.004337310791015625,
0.048492431640625,
0.007625579833984375,
0.00418853759765625,
0.02972412109375,
0.04534912109375,
0.048614501953125,
-0.00698089599609375,
0.09088134765625,
-0.02294921875,
0.04150390625,
0.059478759765625,
0.0243377685546875,
0.06512451171875,
0.0350341796875,
-0.02728271484375,
0.053863525390625,
0.06488037109375,
-0.01023101806640625,
0.057037353515625,
0.00821685791015625,
0.01049041748046875,
-0.0208282470703125,
0.002780914306640625,
-0.047607421875,
0.0234375,
0.013397216796875,
-0.049285888671875,
-0.014404296875,
0.0036144256591796875,
0.006999969482421875,
-0.01120758056640625,
-0.01287841796875,
0.035369873046875,
0.03558349609375,
-0.03265380859375,
0.07135009765625,
0.008575439453125,
0.05572509765625,
-0.052154541015625,
0.0111083984375,
-0.015472412109375,
0.0297393798828125,
-0.0252685546875,
-0.04241943359375,
0.0091705322265625,
-0.007476806640625,
-0.0254364013671875,
-0.01233673095703125,
0.04296875,
-0.04351806640625,
-0.0338134765625,
0.02728271484375,
0.042144775390625,
0.019927978515625,
-0.020599365234375,
-0.0792236328125,
0.00539398193359375,
0.0018444061279296875,
-0.032440185546875,
0.033905029296875,
0.01384735107421875,
0.019195556640625,
0.037811279296875,
0.03759765625,
-0.01052093505859375,
-0.0015697479248046875,
0.016448974609375,
0.059906005859375,
-0.0482177734375,
-0.04296875,
-0.062347412109375,
0.0309295654296875,
-0.0176239013671875,
-0.0269012451171875,
0.063720703125,
0.05120849609375,
0.059539794921875,
-0.01367950439453125,
0.037078857421875,
-0.005382537841796875,
0.0106964111328125,
-0.042877197265625,
0.047760009765625,
-0.050537109375,
-0.002071380615234375,
-0.0198822021484375,
-0.07586669921875,
-0.0137481689453125,
0.0643310546875,
-0.035491943359375,
0.01155853271484375,
0.07244873046875,
0.059814453125,
-0.0157623291015625,
-0.01033782958984375,
0.015899658203125,
0.04022216796875,
0.022125244140625,
0.0609130859375,
0.040374755859375,
-0.08685302734375,
0.0565185546875,
-0.01367950439453125,
-0.01561737060546875,
-0.01506805419921875,
-0.057037353515625,
-0.063720703125,
-0.047088623046875,
-0.045654296875,
-0.0312347412109375,
0.01255035400390625,
0.074462890625,
0.0576171875,
-0.049560546875,
-0.00942230224609375,
0.00024962425231933594,
-0.013916015625,
-0.02825927734375,
-0.0183563232421875,
0.0435791015625,
-0.028564453125,
-0.06903076171875,
0.016754150390625,
-0.01403045654296875,
0.006000518798828125,
0.01094818115234375,
-0.0077056884765625,
-0.048370361328125,
-0.003936767578125,
0.051177978515625,
-0.007843017578125,
-0.0290069580078125,
-0.0281524658203125,
0.0027332305908203125,
-0.0291290283203125,
0.01506805419921875,
0.00981903076171875,
-0.049652099609375,
0.0200653076171875,
0.0518798828125,
0.032623291015625,
0.07293701171875,
0.0002505779266357422,
0.032867431640625,
-0.0545654296875,
0.01029205322265625,
0.00913238525390625,
0.0264129638671875,
0.042144775390625,
-0.0213623046875,
0.036468505859375,
0.032073974609375,
-0.042510986328125,
-0.045654296875,
-0.00861358642578125,
-0.0733642578125,
-0.0162353515625,
0.07733154296875,
-0.01384735107421875,
-0.02410888671875,
0.0091705322265625,
-0.005428314208984375,
0.0291900634765625,
-0.021942138671875,
0.053314208984375,
0.061737060546875,
-0.0120391845703125,
-0.0069427490234375,
-0.0535888671875,
0.040374755859375,
0.038818359375,
-0.037750244140625,
-0.0287933349609375,
0.004207611083984375,
0.03656005859375,
0.0123138427734375,
0.040863037109375,
-0.01267242431640625,
0.003200531005859375,
0.0260162353515625,
-0.0030651092529296875,
-0.005382537841796875,
-0.0093994140625,
-0.0060882568359375,
0.0139312744140625,
-0.020294189453125,
-0.02655029296875
]
] |
timm/vit_base_patch16_224.orig_in21k_ft_in1k | 2023-05-06T00:00:51.000Z | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"dataset:imagenet-1k",
"dataset:imagenet-21k",
"arxiv:2010.11929",
"license:apache-2.0",
"region:us"
] | image-classification | timm | null | null | timm/vit_base_patch16_224.orig_in21k_ft_in1k | 1 | 7,657 | timm | 2022-12-22T07:26:58 | ---
tags:
- image-classification
- timm
library_name: timm
license: apache-2.0
datasets:
- imagenet-1k
- imagenet-21k
---
# Model card for vit_base_patch16_224.orig_in21k_ft_in1k
A Vision Transformer (ViT) image classification model. Trained on ImageNet-21k and fine-tuned on ImageNet-1k in JAX by paper authors, ported to PyTorch by Ross Wightman.
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 86.6
- GMACs: 16.9
- Activations (M): 16.5
- Image size: 224 x 224
- **Papers:**
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale: https://arxiv.org/abs/2010.11929v2
- **Dataset:** ImageNet-1k
- **Pretrain Dataset:** ImageNet-21k
- **Original:** https://github.com/google-research/vision_transformer
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('vit_base_patch16_224.orig_in21k_ft_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'vit_base_patch16_224.orig_in21k_ft_in1k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 197, 768) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
## Citation
```bibtex
@article{dosovitskiy2020vit,
title={An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale},
author={Dosovitskiy, Alexey and Beyer, Lucas and Kolesnikov, Alexander and Weissenborn, Dirk and Zhai, Xiaohua and Unterthiner, Thomas and Dehghani, Mostafa and Minderer, Matthias and Heigold, Georg and Gelly, Sylvain and Uszkoreit, Jakob and Houlsby, Neil},
journal={ICLR},
year={2021}
}
```
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}
```
| 3,394 | [
[
-0.035430908203125,
-0.0299835205078125,
0.0004820823669433594,
0.0101318359375,
-0.028656005859375,
-0.0247955322265625,
-0.01739501953125,
-0.037139892578125,
0.014923095703125,
0.0261077880859375,
-0.037322998046875,
-0.044677734375,
-0.052215576171875,
-0.004367828369140625,
-0.01386260986328125,
0.077880859375,
-0.0103912353515625,
-0.002689361572265625,
-0.0150146484375,
-0.038299560546875,
-0.0120086669921875,
-0.0203704833984375,
-0.04937744140625,
-0.03643798828125,
0.0269317626953125,
0.0117034912109375,
0.048004150390625,
0.04388427734375,
0.057098388671875,
0.03253173828125,
-0.01291656494140625,
-0.001506805419921875,
-0.0197296142578125,
-0.01751708984375,
0.02197265625,
-0.045745849609375,
-0.037811279296875,
0.0196075439453125,
0.053985595703125,
0.02813720703125,
0.00408172607421875,
0.0294647216796875,
0.0092620849609375,
0.040618896484375,
-0.020416259765625,
0.01081085205078125,
-0.0347900390625,
0.01666259765625,
-0.01108551025390625,
0.0047454833984375,
-0.0258026123046875,
-0.0262298583984375,
0.022064208984375,
-0.03692626953125,
0.038665771484375,
-0.0028228759765625,
0.1033935546875,
0.0134735107421875,
-0.00024366378784179688,
0.006351470947265625,
-0.0228118896484375,
0.0550537109375,
-0.054656982421875,
0.02960205078125,
0.0142822265625,
0.01171112060546875,
-0.002880096435546875,
-0.07421875,
-0.0469970703125,
-0.01348114013671875,
-0.01739501953125,
0.002208709716796875,
-0.0301055908203125,
0.01018524169921875,
0.03192138671875,
0.03460693359375,
-0.041351318359375,
-0.0022373199462890625,
-0.038818359375,
-0.018707275390625,
0.038909912109375,
-0.00042176246643066406,
0.01849365234375,
-0.018951416015625,
-0.043792724609375,
-0.04119873046875,
-0.02703857421875,
0.0191497802734375,
0.0233917236328125,
0.007579803466796875,
-0.041839599609375,
0.04144287109375,
0.00875091552734375,
0.0447998046875,
0.014556884765625,
-0.020782470703125,
0.048004150390625,
-0.004505157470703125,
-0.031646728515625,
-0.01519012451171875,
0.08551025390625,
0.032440185546875,
0.0234222412109375,
-0.0009889602661132812,
-0.01091766357421875,
-0.01073455810546875,
-0.00021028518676757812,
-0.0875244140625,
-0.0296630859375,
0.012969970703125,
-0.044464111328125,
-0.0301666259765625,
0.0262908935546875,
-0.052520751953125,
-0.01160430908203125,
-0.00031757354736328125,
0.055694580078125,
-0.0362548828125,
-0.0271453857421875,
0.008331298828125,
-0.01352691650390625,
0.02972412109375,
0.0151519775390625,
-0.036376953125,
0.01141357421875,
0.0194549560546875,
0.08416748046875,
0.003261566162109375,
-0.03497314453125,
-0.01200103759765625,
-0.03375244140625,
-0.02099609375,
0.03802490234375,
-0.003253936767578125,
-0.0166778564453125,
-0.0162506103515625,
0.027923583984375,
-0.01336669921875,
-0.046844482421875,
0.02154541015625,
-0.0147857666015625,
0.023345947265625,
0.004413604736328125,
-0.0149383544921875,
-0.02984619140625,
0.020599365234375,
-0.03619384765625,
0.094482421875,
0.034393310546875,
-0.0689697265625,
0.032440185546875,
-0.040008544921875,
-0.00412750244140625,
-0.006603240966796875,
0.004566192626953125,
-0.08544921875,
-0.0003390312194824219,
0.014923095703125,
0.046478271484375,
-0.0182037353515625,
0.00012445449829101562,
-0.036773681640625,
-0.02001953125,
0.0267333984375,
-0.01245880126953125,
0.0755615234375,
0.005924224853515625,
-0.028167724609375,
0.022735595703125,
-0.041595458984375,
0.00789642333984375,
0.036712646484375,
-0.020111083984375,
-0.0012760162353515625,
-0.0469970703125,
0.019287109375,
0.018890380859375,
0.0125579833984375,
-0.054107666015625,
0.0291748046875,
-0.0178985595703125,
0.03448486328125,
0.045196533203125,
-0.015716552734375,
0.0289306640625,
-0.027099609375,
0.03094482421875,
0.0192108154296875,
0.0233001708984375,
-0.00208282470703125,
-0.04791259765625,
-0.065673828125,
-0.04217529296875,
0.024993896484375,
0.028106689453125,
-0.04132080078125,
0.037506103515625,
-0.02734375,
-0.059722900390625,
-0.03955078125,
-0.0003552436828613281,
0.0379638671875,
0.038665771484375,
0.032379150390625,
-0.038970947265625,
-0.040802001953125,
-0.06561279296875,
-0.00670623779296875,
-0.01018524169921875,
-0.00180816650390625,
0.021728515625,
0.05157470703125,
-0.017974853515625,
0.06182861328125,
-0.03082275390625,
-0.02508544921875,
-0.0184478759765625,
0.00705718994140625,
0.032318115234375,
0.05218505859375,
0.059051513671875,
-0.0457763671875,
-0.04046630859375,
-0.0107574462890625,
-0.064453125,
0.0125579833984375,
0.0031375885009765625,
-0.018096923828125,
0.02191162109375,
0.013671875,
-0.058624267578125,
0.059906005859375,
0.0203857421875,
-0.03802490234375,
0.038055419921875,
-0.0158843994140625,
0.008209228515625,
-0.08428955078125,
0.006160736083984375,
0.027252197265625,
-0.0195770263671875,
-0.033599853515625,
0.00139617919921875,
0.00782012939453125,
-0.0016269683837890625,
-0.0310821533203125,
0.049468994140625,
-0.03546142578125,
-0.003787994384765625,
-0.004833221435546875,
-0.0259552001953125,
0.001132965087890625,
0.049041748046875,
-0.003307342529296875,
0.0338134765625,
0.055023193359375,
-0.030181884765625,
0.0396728515625,
0.037017822265625,
-0.019805908203125,
0.040313720703125,
-0.053985595703125,
0.0135498046875,
0.00267791748046875,
0.0226898193359375,
-0.08001708984375,
-0.01346588134765625,
0.027008056640625,
-0.04840087890625,
0.04400634765625,
-0.0428466796875,
-0.03558349609375,
-0.045745849609375,
-0.03240966796875,
0.036529541015625,
0.0521240234375,
-0.058868408203125,
0.046051025390625,
0.0130767822265625,
0.0216827392578125,
-0.036834716796875,
-0.0679931640625,
-0.0191497802734375,
-0.03424072265625,
-0.054779052734375,
0.035003662109375,
0.00746917724609375,
0.01026153564453125,
0.0095062255859375,
-0.01016998291015625,
0.00316619873046875,
-0.01849365234375,
0.03692626953125,
0.03460693359375,
-0.019317626953125,
-0.0081939697265625,
-0.030426025390625,
-0.01264190673828125,
0.007480621337890625,
-0.025665283203125,
0.041778564453125,
-0.0243072509765625,
-0.00879669189453125,
-0.0535888671875,
-0.0162811279296875,
0.04461669921875,
-0.0161590576171875,
0.060150146484375,
0.080322265625,
-0.034393310546875,
0.0008721351623535156,
-0.037322998046875,
-0.0250396728515625,
-0.036163330078125,
0.03656005859375,
-0.0294647216796875,
-0.032684326171875,
0.06243896484375,
0.0120391845703125,
0.0024566650390625,
0.054595947265625,
0.0302734375,
-0.001178741455078125,
0.060882568359375,
0.046539306640625,
0.0076904296875,
0.0625,
-0.0692138671875,
-0.00785064697265625,
-0.06549072265625,
-0.0306243896484375,
-0.023529052734375,
-0.0404052734375,
-0.055816650390625,
-0.0287322998046875,
0.034393310546875,
0.005107879638671875,
-0.02655029296875,
0.043243408203125,
-0.067626953125,
0.00858306884765625,
0.0546875,
0.041839599609375,
-0.01445770263671875,
0.0277557373046875,
-0.0185546875,
0.0000032186508178710938,
-0.0540771484375,
-0.0123443603515625,
0.08746337890625,
0.036285400390625,
0.056396484375,
-0.009307861328125,
0.045257568359375,
-0.0198974609375,
0.0282440185546875,
-0.05499267578125,
0.04400634765625,
-0.005168914794921875,
-0.03558349609375,
-0.01079559326171875,
-0.0303497314453125,
-0.07952880859375,
0.0113677978515625,
-0.022796630859375,
-0.06280517578125,
0.0174713134765625,
0.01377105712890625,
-0.0149688720703125,
0.055938720703125,
-0.05792236328125,
0.07513427734375,
-0.0009427070617675781,
-0.03033447265625,
0.006435394287109375,
-0.05126953125,
0.0109100341796875,
0.02032470703125,
-0.023223876953125,
0.00296783447265625,
0.020965576171875,
0.08251953125,
-0.04388427734375,
0.056640625,
-0.0309906005859375,
0.02349853515625,
0.03564453125,
-0.00650787353515625,
0.0255889892578125,
-0.004352569580078125,
0.005008697509765625,
0.031219482421875,
0.01200103759765625,
-0.02984619140625,
-0.03680419921875,
0.03851318359375,
-0.0758056640625,
-0.0237274169921875,
-0.040191650390625,
-0.04132080078125,
0.01357269287109375,
0.005954742431640625,
0.051483154296875,
0.04754638671875,
0.0168609619140625,
0.0301513671875,
0.04364013671875,
-0.0241546630859375,
0.02960205078125,
0.0005464553833007812,
-0.0236968994140625,
-0.040679931640625,
0.06365966796875,
0.0178680419921875,
0.01026153564453125,
0.00972747802734375,
0.02117919921875,
-0.029205322265625,
-0.03533935546875,
-0.0266571044921875,
0.03582763671875,
-0.049102783203125,
-0.0396728515625,
-0.0423583984375,
-0.03472900390625,
-0.025482177734375,
0.001087188720703125,
-0.03350830078125,
-0.021881103515625,
-0.0215911865234375,
0.005985260009765625,
0.06353759765625,
0.044189453125,
-0.006084442138671875,
0.037811279296875,
-0.03955078125,
0.01045989990234375,
0.01396942138671875,
0.043182373046875,
-0.01325225830078125,
-0.0755615234375,
-0.021392822265625,
-0.0037784576416015625,
-0.037017822265625,
-0.059661865234375,
0.03692626953125,
0.01090240478515625,
0.030670166015625,
0.023895263671875,
-0.0186920166015625,
0.06134033203125,
-0.00495147705078125,
0.040008544921875,
0.02508544921875,
-0.04718017578125,
0.0390625,
-0.00980377197265625,
0.01557159423828125,
0.007778167724609375,
0.0171966552734375,
-0.02294921875,
-0.00445556640625,
-0.08038330078125,
-0.058807373046875,
0.06298828125,
0.0167388916015625,
0.0017156600952148438,
0.03131103515625,
0.047607421875,
-0.002288818359375,
0.0019702911376953125,
-0.06231689453125,
-0.0268096923828125,
-0.029632568359375,
-0.0255889892578125,
-0.0028362274169921875,
-0.005367279052734375,
-0.001483917236328125,
-0.053985595703125,
0.05157470703125,
-0.005962371826171875,
0.0628662109375,
0.0311431884765625,
-0.011016845703125,
-0.01380157470703125,
-0.0261993408203125,
0.0247802734375,
0.0198974609375,
-0.029205322265625,
0.00441741943359375,
0.01332855224609375,
-0.05419921875,
-0.0018072128295898438,
0.019683837890625,
0.00018107891082763672,
0.004405975341796875,
0.03570556640625,
0.07586669921875,
-0.0059814453125,
0.0002703666687011719,
0.0350341796875,
-0.0049896240234375,
-0.03228759765625,
-0.0190277099609375,
0.0014514923095703125,
-0.01010894775390625,
0.0289154052734375,
0.022491455078125,
0.0262298583984375,
-0.0131988525390625,
-0.0160980224609375,
0.01364898681640625,
0.046478271484375,
-0.03350830078125,
-0.03192138671875,
0.046661376953125,
-0.020599365234375,
-0.01261138916015625,
0.06640625,
-0.0024127960205078125,
-0.04205322265625,
0.07122802734375,
0.0263519287109375,
0.07269287109375,
-0.007434844970703125,
0.0016536712646484375,
0.0589599609375,
0.0186614990234375,
-0.003551483154296875,
0.01497650146484375,
0.01282501220703125,
-0.06341552734375,
0.006336212158203125,
-0.046417236328125,
0.0006394386291503906,
0.0276947021484375,
-0.04522705078125,
0.0279388427734375,
-0.044891357421875,
-0.0347900390625,
0.0116119384765625,
0.0197906494140625,
-0.07281494140625,
0.02099609375,
0.00131988525390625,
0.06024169921875,
-0.05975341796875,
0.0555419921875,
0.067138671875,
-0.04681396484375,
-0.0760498046875,
-0.01224517822265625,
-0.0023937225341796875,
-0.0611572265625,
0.037933349609375,
0.032501220703125,
0.0088958740234375,
0.0159454345703125,
-0.0611572265625,
-0.0467529296875,
0.1004638671875,
0.0330810546875,
-0.0054931640625,
0.01346588134765625,
-0.0026035308837890625,
0.0254364013671875,
-0.023529052734375,
0.041046142578125,
0.017791748046875,
0.028167724609375,
0.0190582275390625,
-0.059661865234375,
0.013092041015625,
-0.024810791015625,
0.00876617431640625,
0.0180206298828125,
-0.05792236328125,
0.0726318359375,
-0.0295562744140625,
-0.009124755859375,
0.01605224609375,
0.047698974609375,
0.016693115234375,
0.00518035888671875,
0.042816162109375,
0.066162109375,
0.0307159423828125,
-0.03094482421875,
0.06744384765625,
-0.00537872314453125,
0.059295654296875,
0.037811279296875,
0.03302001953125,
0.041046142578125,
0.036834716796875,
-0.033782958984375,
0.034637451171875,
0.0709228515625,
-0.034210205078125,
0.022705078125,
0.009796142578125,
0.0023021697998046875,
-0.0132598876953125,
0.00350189208984375,
-0.031829833984375,
0.038177490234375,
0.01342010498046875,
-0.041351318359375,
-0.0110321044921875,
0.00634765625,
0.000774383544921875,
-0.0303497314453125,
-0.01280975341796875,
0.03936767578125,
0.000002205371856689453,
-0.032012939453125,
0.058441162109375,
0.0005726814270019531,
0.0645751953125,
-0.034515380859375,
-0.0002446174621582031,
-0.0191192626953125,
0.0325927734375,
-0.0283203125,
-0.0653076171875,
0.018707275390625,
-0.016265869140625,
-0.004596710205078125,
-0.001827239990234375,
0.050628662109375,
-0.0277099609375,
-0.041259765625,
0.0156707763671875,
0.018463134765625,
0.0282745361328125,
-0.00978851318359375,
-0.08294677734375,
0.0033416748046875,
-0.0013189315795898438,
-0.047088623046875,
0.021331787109375,
0.0328369140625,
0.002475738525390625,
0.053192138671875,
0.0458984375,
-0.00855255126953125,
0.01364898681640625,
-0.01322174072265625,
0.06378173828125,
-0.034332275390625,
-0.0271453857421875,
-0.062042236328125,
0.047637939453125,
0.0011119842529296875,
-0.046630859375,
0.04681396484375,
0.042388916015625,
0.0679931640625,
-0.00803375244140625,
0.032073974609375,
-0.02154541015625,
0.0010576248168945312,
-0.0233001708984375,
0.054412841796875,
-0.05584716796875,
-0.011688232421875,
-0.0267486572265625,
-0.06134033203125,
-0.0304718017578125,
0.0689697265625,
-0.0223846435546875,
0.03411865234375,
0.0426025390625,
0.07769775390625,
-0.0260009765625,
-0.025115966796875,
0.0142059326171875,
0.01739501953125,
0.009857177734375,
0.0302276611328125,
0.037109375,
-0.06304931640625,
0.0347900390625,
-0.045440673828125,
-0.014373779296875,
-0.0145721435546875,
-0.047607421875,
-0.07745361328125,
-0.06365966796875,
-0.046356201171875,
-0.060699462890625,
-0.0196533203125,
0.07012939453125,
0.07623291015625,
-0.050811767578125,
-0.01035308837890625,
-0.00855255126953125,
0.0006833076477050781,
-0.0208282470703125,
-0.0187530517578125,
0.043853759765625,
-0.0011444091796875,
-0.05633544921875,
-0.0311431884765625,
-0.002155303955078125,
0.0304412841796875,
-0.00848388671875,
-0.0157623291015625,
-0.00980377197265625,
-0.0218658447265625,
0.0163421630859375,
0.022796630859375,
-0.053985595703125,
-0.01739501953125,
-0.0080108642578125,
-0.0083465576171875,
0.037017822265625,
0.02874755859375,
-0.05181884765625,
0.031707763671875,
0.04046630859375,
0.0202789306640625,
0.06475830078125,
-0.01885986328125,
0.0023212432861328125,
-0.055450439453125,
0.041046142578125,
-0.005817413330078125,
0.03961181640625,
0.033935546875,
-0.022125244140625,
0.043426513671875,
0.042144775390625,
-0.0338134765625,
-0.0634765625,
-0.004055023193359375,
-0.0831298828125,
0.00507354736328125,
0.0733642578125,
-0.0278167724609375,
-0.03643798828125,
0.0333251953125,
-0.0117645263671875,
0.048583984375,
-0.0093994140625,
0.033935546875,
0.0249481201171875,
-0.0004036426544189453,
-0.04644775390625,
-0.0294036865234375,
0.034088134765625,
0.0130157470703125,
-0.040802001953125,
-0.028717041015625,
-0.0006899833679199219,
0.050323486328125,
0.025787353515625,
0.0283050537109375,
-0.0166778564453125,
0.0113677978515625,
0.003204345703125,
0.04266357421875,
-0.0228271484375,
-0.010467529296875,
-0.0272216796875,
-0.005222320556640625,
-0.00701904296875,
-0.043243408203125
]
] |
meta-math/MetaMath-70B-V1.0 | 2023-10-11T02:44:28.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:meta-math/MetaMathQA",
"arxiv:2309.12284",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us",
"has_space"
] | text-generation | meta-math | null | null | meta-math/MetaMath-70B-V1.0 | 10 | 7,657 | transformers | 2023-09-22T03:20:33 | ---
license: llama2
datasets:
- meta-math/MetaMathQA
---
arxiv.org/abs/2309.12284
View the project page:
https://meta-math.github.io/
# Citation
```bibtex
@article{yu2023metamath,
title={MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models},
author={Yu, Longhui and Jiang, Weisen and Shi, Han and Yu, Jincheng and Liu, Zhengying and Zhang, Yu and Kwok, James T and Li, Zhenguo and Weller, Adrian and Liu, Weiyang},
journal={arXiv preprint arXiv:2309.12284},
year={2023}
}
``` | 512 | [
[
-0.0313720703125,
-0.0390625,
0.048980712890625,
0.0186614990234375,
0.00518035888671875,
-0.011260986328125,
-0.017120361328125,
-0.018585205078125,
0.046630859375,
0.0171966552734375,
-0.042205810546875,
-0.02197265625,
-0.0150604248046875,
0.0095672607421875,
-0.00882720947265625,
0.081787109375,
-0.003398895263671875,
0.01317596435546875,
-0.0226898193359375,
-0.02886962890625,
0.01617431640625,
-0.03155517578125,
-0.025177001953125,
-0.0002503395080566406,
0.034393310546875,
0.0140228271484375,
0.0474853515625,
0.0572509765625,
0.033203125,
0.021209716796875,
0.01448822021484375,
-0.0308074951171875,
-0.0006422996520996094,
0.00514984130859375,
-0.01482391357421875,
-0.005802154541015625,
-0.061798095703125,
0.008544921875,
0.06329345703125,
0.06744384765625,
-0.0048675537109375,
0.046966552734375,
0.0130462646484375,
0.0379638671875,
-0.041046142578125,
0.016448974609375,
-0.031219482421875,
-0.0419921875,
-0.0298614501953125,
-0.0035228729248046875,
-0.030120849609375,
-0.019561767578125,
0.0011224746704101562,
-0.04290771484375,
0.01477813720703125,
-0.0081939697265625,
0.06842041015625,
0.01351165771484375,
-0.0216217041015625,
0.0082244873046875,
-0.031646728515625,
0.06671142578125,
-0.056182861328125,
0.0279541015625,
0.049224853515625,
0.02008056640625,
0.005382537841796875,
-0.058502197265625,
-0.0237579345703125,
0.01971435546875,
-0.0233612060546875,
-0.01389312744140625,
-0.010528564453125,
-0.0156402587890625,
0.0250396728515625,
0.0131072998046875,
-0.05755615234375,
-0.0296478271484375,
-0.038665771484375,
-0.0217132568359375,
0.051727294921875,
0.01514434814453125,
0.010528564453125,
-0.038330078125,
-0.0198211669921875,
-0.00569915771484375,
-0.042877197265625,
0.0092926025390625,
0.018096923828125,
0.013916015625,
-0.03729248046875,
0.036712646484375,
0.00345611572265625,
0.0494384765625,
-0.00945281982421875,
-0.0031871795654296875,
0.0643310546875,
-0.049560546875,
-0.0019292831420898438,
-0.042266845703125,
0.0972900390625,
0.0007090568542480469,
0.005878448486328125,
0.0174407958984375,
0.00144195556640625,
-0.052459716796875,
-0.00618743896484375,
-0.07086181640625,
-0.0164337158203125,
0.022796630859375,
-0.016510009765625,
-0.0062255859375,
0.03338623046875,
-0.088623046875,
-0.003185272216796875,
-0.0172882080078125,
0.0234832763671875,
0.00640869140625,
-0.03265380859375,
-0.01497650146484375,
0.0159759521484375,
0.020721435546875,
-0.0035381317138671875,
-0.06414794921875,
0.0191497802734375,
0.038665771484375,
0.0484619140625,
0.011199951171875,
-0.0236663818359375,
-0.054290771484375,
-0.004505157470703125,
-0.0111846923828125,
0.02838134765625,
-0.049224853515625,
0.0084228515625,
0.0091552734375,
0.0259552001953125,
-0.0097808837890625,
-0.0238494873046875,
0.037353515625,
-0.040740966796875,
0.032928466796875,
-0.0302581787109375,
0.004535675048828125,
-0.005126953125,
0.0157470703125,
-0.056854248046875,
0.06744384765625,
0.0153656005859375,
-0.05828857421875,
-0.00592803955078125,
-0.045257568359375,
-0.0160369873046875,
0.00644683837890625,
-0.005695343017578125,
-0.04180908203125,
-0.0169525146484375,
0.0160980224609375,
0.0179443359375,
-0.0288238525390625,
0.03961181640625,
-0.0386962890625,
0.0007262229919433594,
0.016021728515625,
-0.01029205322265625,
0.08642578125,
0.0249176025390625,
-0.0108489990234375,
0.0167388916015625,
-0.0745849609375,
0.0128173828125,
0.022064208984375,
-0.02001953125,
-0.043212890625,
-0.000278472900390625,
0.007251739501953125,
0.0080108642578125,
0.0389404296875,
-0.024749755859375,
0.02703857421875,
-0.00461578369140625,
0.0400390625,
0.03851318359375,
-0.012786865234375,
0.027130126953125,
-0.00470733642578125,
0.046051025390625,
-0.0172119140625,
-0.0118560791015625,
-0.044219970703125,
-0.0292510986328125,
-0.05712890625,
-0.0250091552734375,
0.05609130859375,
0.037933349609375,
-0.046722412109375,
0.0394287109375,
-0.0291900634765625,
-0.024658203125,
-0.0498046875,
-0.0037708282470703125,
0.01947021484375,
0.030609130859375,
0.052703857421875,
0.01348876953125,
-0.050811767578125,
-0.07086181640625,
-0.018157958984375,
-0.0187835693359375,
-0.0036258697509765625,
0.028045654296875,
0.060150146484375,
-0.0241546630859375,
0.07470703125,
-0.05419921875,
0.01300048828125,
0.005615234375,
0.0280914306640625,
0.0173187255859375,
0.040557861328125,
0.04351806640625,
-0.031951904296875,
-0.06695556640625,
-0.014892578125,
-0.042327880859375,
-0.044281005859375,
-0.00112152099609375,
-0.017486572265625,
0.033843994140625,
0.0280914306640625,
-0.047698974609375,
0.027496337890625,
0.031646728515625,
-0.04644775390625,
0.06121826171875,
0.0211639404296875,
0.0189208984375,
-0.108154296875,
0.05303955078125,
-0.0081634521484375,
-0.028228759765625,
-0.0311126708984375,
0.0166778564453125,
0.0162811279296875,
0.0004699230194091797,
-0.0144500732421875,
0.05059814453125,
-0.041473388671875,
0.0073699951171875,
0.0036640167236328125,
-0.003993988037109375,
-0.003971099853515625,
0.03607177734375,
-0.01511383056640625,
0.059722900390625,
0.045806884765625,
-0.036956787109375,
0.0291900634765625,
0.01309967041015625,
-0.023284912109375,
0.031829833984375,
-0.0782470703125,
0.0021724700927734375,
0.018096923828125,
0.0305633544921875,
-0.07196044921875,
-0.0198516845703125,
0.032684326171875,
-0.04156494140625,
0.005008697509765625,
-0.0217132568359375,
-0.04931640625,
-0.0191192626953125,
-0.041717529296875,
0.074462890625,
0.0305328369140625,
-0.0174407958984375,
0.02783203125,
0.022247314453125,
-0.0244903564453125,
-0.04638671875,
-0.037445068359375,
-0.016021728515625,
-0.022247314453125,
-0.05108642578125,
0.0157470703125,
-0.03765869140625,
-0.034576416015625,
0.00035881996154785156,
0.0177764892578125,
-0.0026149749755859375,
-0.0193023681640625,
-0.01361083984375,
0.0239105224609375,
-0.038299560546875,
0.021209716796875,
-0.0022735595703125,
-0.036956787109375,
0.039276123046875,
-0.0181732177734375,
0.0631103515625,
-0.0096282958984375,
-0.019134521484375,
-0.019256591796875,
0.0241851806640625,
0.053253173828125,
-0.04022216796875,
0.0604248046875,
0.04766845703125,
-0.035308837890625,
0.022796630859375,
-0.038177490234375,
-0.0031032562255859375,
-0.0367431640625,
0.0232696533203125,
-0.0176239013671875,
-0.03216552734375,
0.05828857421875,
0.01399993896484375,
0.01096343994140625,
0.07818603515625,
0.046112060546875,
0.00830841064453125,
0.049591064453125,
0.0269775390625,
0.0067596435546875,
0.0222320556640625,
-0.0166778564453125,
-0.0150604248046875,
-0.07525634765625,
-0.00937652587890625,
-0.0487060546875,
-0.0211029052734375,
-0.0364990234375,
-0.042327880859375,
0.0261383056640625,
-0.00370025634765625,
-0.047698974609375,
0.0577392578125,
-0.01120758056640625,
0.0291900634765625,
0.047607421875,
-0.0157623291015625,
0.018096923828125,
-0.01171875,
-0.025421142578125,
-0.027191162109375,
-0.01230621337890625,
-0.0212860107421875,
0.07666015625,
0.0322265625,
0.046844482421875,
0.0271148681640625,
0.047088623046875,
-0.0253143310546875,
0.0021686553955078125,
-0.025634765625,
0.03302001953125,
0.034942626953125,
-0.08673095703125,
-0.0308380126953125,
-0.051513671875,
-0.07958984375,
0.0128326416015625,
-0.0038814544677734375,
-0.045867919921875,
0.0103607177734375,
-0.01274871826171875,
-0.0254058837890625,
0.0164794921875,
-0.03668212890625,
0.06341552734375,
0.0021800994873046875,
-0.04052734375,
0.0008387565612792969,
-0.061798095703125,
0.0226593017578125,
-0.0138702392578125,
0.04254150390625,
0.007122039794921875,
-0.010162353515625,
0.0780029296875,
-0.039093017578125,
0.0467529296875,
-0.0019989013671875,
-0.0063629150390625,
0.0303497314453125,
0.006015777587890625,
0.0250091552734375,
0.026824951171875,
-0.007965087890625,
0.0028629302978515625,
0.01055145263671875,
-0.03851318359375,
-0.034698486328125,
0.054595947265625,
-0.050079345703125,
-0.039947509765625,
-0.0601806640625,
-0.042388916015625,
-0.002857208251953125,
0.043609619140625,
0.018463134765625,
0.03228759765625,
-0.015869140625,
0.04376220703125,
0.0281524658203125,
-0.006595611572265625,
0.06182861328125,
0.04547119140625,
-0.030609130859375,
-0.061614990234375,
0.048187255859375,
0.01904296875,
0.018829345703125,
0.0247650146484375,
0.0289154052734375,
-0.0011186599731445312,
-0.0002435445785522461,
-0.04400634765625,
0.0548095703125,
-0.0273895263671875,
-0.030517578125,
-0.044464111328125,
-0.034027099609375,
-0.0303955078125,
0.00492095947265625,
-0.0447998046875,
-0.0362548828125,
-0.03155517578125,
0.01360321044921875,
0.02484130859375,
0.0310516357421875,
-0.0177764892578125,
0.0015468597412109375,
-0.048004150390625,
0.0120849609375,
0.0201568603515625,
0.0421142578125,
0.006450653076171875,
-0.057403564453125,
-0.048675537109375,
0.01357269287109375,
-0.0328369140625,
-0.049652099609375,
0.0268707275390625,
-0.00754547119140625,
0.0498046875,
0.0226287841796875,
0.0260162353515625,
0.047607421875,
-0.0467529296875,
0.05645751953125,
0.02069091796875,
-0.07318115234375,
0.032928466796875,
-0.043426513671875,
0.033477783203125,
0.0556640625,
0.041107177734375,
-0.01424407958984375,
0.00739288330078125,
-0.060882568359375,
-0.06854248046875,
0.039154052734375,
0.01284027099609375,
0.0093536376953125,
0.016998291015625,
0.006595611572265625,
0.003299713134765625,
0.005657196044921875,
-0.09185791015625,
-0.05047607421875,
-0.00324249267578125,
-0.0256195068359375,
-0.00539398193359375,
-0.032440185546875,
-0.043701171875,
-0.042388916015625,
0.044219970703125,
-0.01056671142578125,
0.0296478271484375,
-0.003147125244140625,
-0.011749267578125,
-0.024383544921875,
0.0282135009765625,
0.068115234375,
0.060455322265625,
-0.0160675048828125,
0.0106048583984375,
0.00466156005859375,
-0.04266357421875,
-0.01007843017578125,
0.0372314453125,
-0.005191802978515625,
-0.0092620849609375,
0.048095703125,
0.033416748046875,
0.0161895751953125,
-0.0274200439453125,
0.03973388671875,
0.030181884765625,
-0.049224853515625,
-0.0333251953125,
-0.020721435546875,
0.016754150390625,
0.0181121826171875,
0.05584716796875,
-0.00461578369140625,
0.0066680908203125,
-0.030517578125,
0.0159149169921875,
0.0275115966796875,
-0.0164794921875,
-0.052215576171875,
0.043426513671875,
0.00782012939453125,
-0.027587890625,
0.0183868408203125,
-0.0284881591796875,
-0.042236328125,
0.02069091796875,
0.046539306640625,
0.053802490234375,
-0.0084991455078125,
-0.004974365234375,
0.06634521484375,
0.0236663818359375,
0.00951385498046875,
0.01528167724609375,
0.0196075439453125,
-0.02557373046875,
-0.0262298583984375,
-0.031219482421875,
-0.00798797607421875,
0.03662109375,
-0.053741455078125,
0.042755126953125,
-0.0250091552734375,
0.0044708251953125,
-0.006046295166015625,
0.012176513671875,
-0.0283203125,
-0.00424957275390625,
-0.0016069412231445312,
0.0455322265625,
-0.047454833984375,
0.047882080078125,
0.057403564453125,
-0.03961181640625,
-0.040069580078125,
0.01500701904296875,
0.005268096923828125,
-0.043914794921875,
0.017486572265625,
-0.00012600421905517578,
0.003459930419921875,
-0.004108428955078125,
-0.0657958984375,
-0.09320068359375,
0.0860595703125,
0.033050537109375,
-0.029571533203125,
0.01331329345703125,
-0.0131988525390625,
0.0234222412109375,
-0.0100555419921875,
0.025421142578125,
0.0179443359375,
0.05157470703125,
0.0238494873046875,
-0.074462890625,
0.00994873046875,
-0.056427001953125,
-0.040863037109375,
0.0300445556640625,
-0.07281494140625,
0.07159423828125,
-0.019500732421875,
-0.00292205810546875,
0.006717681884765625,
0.06427001953125,
0.0310516357421875,
0.031829833984375,
0.0221405029296875,
0.0557861328125,
0.040771484375,
-0.014678955078125,
0.0253143310546875,
-0.0196380615234375,
0.057464599609375,
0.0787353515625,
0.01445770263671875,
0.0694580078125,
0.04107666015625,
-0.0421142578125,
0.073486328125,
0.032073974609375,
-0.0171051025390625,
0.0408935546875,
0.01111602783203125,
0.010986328125,
-0.0287322998046875,
0.02587890625,
-0.07470703125,
0.01171112060546875,
0.0087432861328125,
-0.03155517578125,
-0.005054473876953125,
-0.028900146484375,
0.033905029296875,
-0.0034770965576171875,
0.00450897216796875,
0.0186767578125,
0.01171875,
-0.03900146484375,
0.04718017578125,
0.00494384765625,
0.036407470703125,
-0.046142578125,
0.00458526611328125,
-0.0020580291748046875,
0.0088348388671875,
-0.0185394287109375,
-0.045501708984375,
0.032684326171875,
0.0049285888671875,
-0.0254058837890625,
0.0026836395263671875,
0.0143280029296875,
-0.033111572265625,
-0.07012939453125,
0.00669097900390625,
0.03472900390625,
0.01259613037109375,
0.015045166015625,
-0.0499267578125,
-0.0067596435546875,
-0.006946563720703125,
-0.050079345703125,
0.0198974609375,
0.0361328125,
0.0162353515625,
0.037567138671875,
0.0478515625,
-0.0248260498046875,
-0.00701141357421875,
-0.013519287109375,
0.060333251953125,
-0.07159423828125,
-0.035888671875,
-0.08837890625,
0.0650634765625,
-0.003917694091796875,
-0.049285888671875,
0.050079345703125,
0.058380126953125,
0.035736083984375,
-0.0016345977783203125,
0.038818359375,
-0.0255279541015625,
0.053680419921875,
-0.0173797607421875,
0.0650634765625,
-0.0595703125,
0.01641845703125,
-0.045501708984375,
-0.07330322265625,
-0.035675048828125,
0.0523681640625,
-0.01861572265625,
0.0323486328125,
0.08197021484375,
0.036407470703125,
-0.01183319091796875,
-0.0330810546875,
0.0024566650390625,
0.0175018310546875,
0.0184478759765625,
0.033843994140625,
0.0185394287109375,
-0.051055908203125,
0.0406494140625,
-0.00043964385986328125,
-0.0203704833984375,
-0.025238037109375,
-0.07806396484375,
-0.06658935546875,
-0.067626953125,
-0.0279998779296875,
-0.045318603515625,
-0.039306640625,
0.07891845703125,
0.045928955078125,
-0.072265625,
-0.018035888671875,
0.014129638671875,
0.0265350341796875,
-0.0018091201782226562,
-0.01934814453125,
0.0435791015625,
-0.005603790283203125,
-0.05938720703125,
0.005603790283203125,
-0.011260986328125,
0.003711700439453125,
-0.0295257568359375,
-0.0161895751953125,
-0.038818359375,
0.006206512451171875,
0.01328277587890625,
0.03009033203125,
-0.0379638671875,
0.004413604736328125,
0.0213775634765625,
-0.0316162109375,
-0.00414276123046875,
0.0418701171875,
-0.042816162109375,
0.00765228271484375,
0.05731201171875,
0.0318603515625,
0.020111083984375,
0.007488250732421875,
0.03863525390625,
-0.036407470703125,
0.0008916854858398438,
-0.004123687744140625,
0.0306396484375,
0.012481689453125,
0.00948333740234375,
0.050201416015625,
0.046234130859375,
-0.0469970703125,
-0.073974609375,
0.00927734375,
-0.088134765625,
-0.0014190673828125,
0.097412109375,
-0.0021800994873046875,
0.0012950897216796875,
0.002475738525390625,
0.00981903076171875,
0.00023853778839111328,
-0.0286712646484375,
0.023712158203125,
0.07159423828125,
0.0338134765625,
-0.031463623046875,
-0.050079345703125,
0.0203704833984375,
0.0101776123046875,
-0.0587158203125,
0.004367828369140625,
0.007411956787109375,
0.0250396728515625,
0.046417236328125,
0.0308380126953125,
-0.018310546875,
0.0208892822265625,
-0.00882720947265625,
0.019073486328125,
-0.01016998291015625,
-0.031829833984375,
-0.01398468017578125,
-0.0029010772705078125,
-0.007793426513671875,
-0.01013946533203125
]
] |
willyninja30/ARIA-70B-French | 2023-10-05T00:45:59.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"code",
"fr",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | willyninja30 | null | null | willyninja30/ARIA-70B-French | 0 | 7,653 | transformers | 2023-09-08T12:54:46 | ---
license: llama2
inference: true
language:
- fr
tags:
- pytorch
- code
---
# ARIA French is a model created from llama 2 70B finetuned on a french dataset.
# contact@faradaylab.fr if you need additional support or integration to your company data.
# Model Developers :FARADAY
# ARIA is the first version of our models based on Llama 2-70B-Chat-HF. We finetuned llama 2 over 10.000 high quality french tokens. This version has been trained on a small dataset extract from the french parliament.
The goal is to increase model quality on French and general topics.
# 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
1 Day using NVIDIA A100 and a cloud service. 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 and accept our License before requesting access here.
# 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/ | 2,448 | [
[
-0.0142974853515625,
-0.06573486328125,
0.02154541015625,
0.052734375,
-0.01215362548828125,
-0.01152801513671875,
-0.01255035400390625,
-0.044189453125,
0.0178680419921875,
0.0533447265625,
-0.0400390625,
-0.0401611328125,
-0.046783447265625,
0.020172119140625,
-0.04290771484375,
0.08123779296875,
-0.00665283203125,
-0.007144927978515625,
0.01399993896484375,
-0.0089111328125,
-0.0105743408203125,
-0.050567626953125,
-0.07037353515625,
-0.0307769775390625,
0.05084228515625,
0.0460205078125,
0.04510498046875,
0.048095703125,
0.0260162353515625,
0.0203704833984375,
-0.0255889892578125,
0.01375579833984375,
-0.042877197265625,
-0.0258941650390625,
0.004138946533203125,
-0.049407958984375,
-0.051177978515625,
-0.00728607177734375,
0.04193115234375,
0.01091766357421875,
-0.004505157470703125,
0.0267791748046875,
0.00907135009765625,
0.051910400390625,
-0.0035877227783203125,
0.02581787109375,
-0.041595458984375,
-0.01085662841796875,
-0.0142059326171875,
-0.0066070556640625,
-0.027618408203125,
-0.00444793701171875,
0.00119781494140625,
-0.05755615234375,
-0.005779266357421875,
0.0180816650390625,
0.087646484375,
0.05194091796875,
-0.038299560546875,
-0.0150146484375,
-0.053009033203125,
0.070556640625,
-0.041717529296875,
0.0170745849609375,
0.0440673828125,
0.0219268798828125,
-0.005962371826171875,
-0.074462890625,
-0.038421630859375,
-0.022552490234375,
0.00913238525390625,
-0.01543426513671875,
-0.03546142578125,
-0.01229095458984375,
0.002422332763671875,
0.030029296875,
-0.036590576171875,
0.0306243896484375,
-0.03497314453125,
-0.01230621337890625,
0.0537109375,
-0.0035457611083984375,
0.019378662109375,
-0.007232666015625,
-0.0338134765625,
-0.0218658447265625,
-0.064208984375,
0.004009246826171875,
0.0304718017578125,
0.032135009765625,
-0.03717041015625,
0.0484619140625,
-0.02130126953125,
0.018646240234375,
0.00585174560546875,
-0.001750946044921875,
0.0293426513671875,
0.01161956787109375,
-0.02642822265625,
-0.0018186569213867188,
0.06671142578125,
0.0218048095703125,
0.033050537109375,
-0.005779266357421875,
-0.01369476318359375,
0.0084381103515625,
0.011016845703125,
-0.054351806640625,
-0.005535125732421875,
0.023681640625,
-0.0265960693359375,
-0.035888671875,
-0.0372314453125,
-0.037628173828125,
-0.0037059783935546875,
-0.0333251953125,
0.0205078125,
-0.01323699951171875,
-0.01450347900390625,
0.011016845703125,
0.032470703125,
0.024139404296875,
0.0164642333984375,
-0.054901123046875,
0.0120849609375,
0.0333251953125,
0.04412841796875,
0.00591278076171875,
-0.02569580078125,
-0.03857421875,
-0.019683837890625,
-0.0235137939453125,
0.05731201171875,
-0.0247039794921875,
-0.01372528076171875,
0.01459503173828125,
0.0300445556640625,
0.012908935546875,
-0.0211029052734375,
0.081298828125,
-0.050323486328125,
0.0097198486328125,
-0.0146026611328125,
-0.027069091796875,
-0.0306549072265625,
0.00530242919921875,
-0.05950927734375,
0.1015625,
0.0009002685546875,
-0.033447265625,
0.01727294921875,
-0.05255126953125,
-0.0148773193359375,
-0.0149688720703125,
0.004024505615234375,
-0.0426025390625,
-0.0166015625,
0.039642333984375,
0.033599853515625,
-0.0265655517578125,
0.01678466796875,
0.007640838623046875,
-0.01904296875,
0.00872802734375,
-0.03887939453125,
0.041412353515625,
0.033599853515625,
-0.02325439453125,
0.0028400421142578125,
-0.0677490234375,
-0.02001953125,
0.0215911865234375,
-0.034576416015625,
-0.010955810546875,
0.003681182861328125,
0.036865234375,
0.0160064697265625,
0.024383544921875,
-0.03289794921875,
0.01184844970703125,
-0.055206298828125,
0.0357666015625,
0.036041259765625,
0.0026264190673828125,
0.032073974609375,
-0.027069091796875,
0.032623291015625,
-0.002826690673828125,
0.0018148422241210938,
0.01007843017578125,
-0.038970947265625,
-0.08740234375,
-0.0121002197265625,
0.027313232421875,
0.05023193359375,
-0.045379638671875,
0.02557373046875,
-0.0201416015625,
-0.038421630859375,
-0.038909912109375,
0.01076507568359375,
0.040283203125,
0.0110931396484375,
0.043731689453125,
-0.00824737548828125,
-0.026123046875,
-0.072265625,
0.005428314208984375,
-0.00713348388671875,
-0.003047943115234375,
0.022247314453125,
0.0531005859375,
-0.027984619140625,
0.051239013671875,
-0.0189208984375,
-0.037933349609375,
-0.0259857177734375,
0.0167694091796875,
0.01552581787109375,
0.0255889892578125,
0.0670166015625,
-0.051971435546875,
-0.023895263671875,
0.006862640380859375,
-0.05682373046875,
-0.005779266357421875,
-0.007419586181640625,
-0.01233673095703125,
0.018157958984375,
0.0252838134765625,
-0.0191802978515625,
0.0283355712890625,
0.06103515625,
-0.0203399658203125,
0.0181427001953125,
0.0023326873779296875,
-0.0042572021484375,
-0.10748291015625,
-0.014923095703125,
0.0160980224609375,
-0.0206146240234375,
-0.039764404296875,
0.0016269683837890625,
-0.00494384765625,
-0.00437164306640625,
-0.0679931640625,
0.0469970703125,
-0.0244903564453125,
0.008758544921875,
-0.01012420654296875,
0.00394439697265625,
0.01157379150390625,
0.067138671875,
0.0027637481689453125,
0.0755615234375,
0.034912109375,
-0.057220458984375,
0.01165771484375,
0.05755615234375,
-0.0247344970703125,
0.0285797119140625,
-0.07366943359375,
0.00783538818359375,
0.004917144775390625,
0.0141448974609375,
-0.0576171875,
-0.020599365234375,
0.017425537109375,
-0.0245819091796875,
0.006397247314453125,
-0.011749267578125,
-0.03314208984375,
-0.0198211669921875,
0.0005512237548828125,
0.032440185546875,
0.040069580078125,
-0.061553955078125,
0.039825439453125,
0.020965576171875,
-0.02410888671875,
-0.061309814453125,
-0.07672119140625,
0.019439697265625,
-0.0293426513671875,
-0.0565185546875,
0.0265655517578125,
0.0068359375,
-0.00958251953125,
-0.0157470703125,
0.01308441162109375,
-0.0188751220703125,
-0.007366180419921875,
0.016693115234375,
0.0014896392822265625,
-0.0211944580078125,
0.021697998046875,
0.0263671875,
-0.004608154296875,
-0.0083465576171875,
-0.0133819580078125,
0.040283203125,
-0.037841796875,
-0.0198516845703125,
-0.0743408203125,
0.00829315185546875,
0.048797607421875,
-0.0288848876953125,
0.057220458984375,
0.029876708984375,
-0.01629638671875,
-0.019378662109375,
-0.043304443359375,
-0.008514404296875,
-0.036865234375,
0.03118896484375,
-0.0288848876953125,
-0.0633544921875,
0.041717529296875,
-0.0092926025390625,
0.0221099853515625,
0.053985595703125,
0.046234130859375,
-0.002742767333984375,
0.057861328125,
0.06622314453125,
-0.020355224609375,
0.0343017578125,
-0.0482177734375,
-0.0028514862060546875,
-0.0574951171875,
-0.01007843017578125,
-0.053192138671875,
-0.0186920166015625,
-0.0362548828125,
-0.031524658203125,
0.0136871337890625,
-0.0022182464599609375,
-0.0318603515625,
0.03704833984375,
-0.04278564453125,
0.0286865234375,
0.020111083984375,
0.007221221923828125,
0.034271240234375,
0.0271759033203125,
0.001987457275390625,
0.00754547119140625,
-0.04412841796875,
-0.0849609375,
0.09521484375,
0.0611572265625,
0.062042236328125,
0.01116943359375,
0.0408935546875,
0.026275634765625,
0.0212554931640625,
-0.055572509765625,
0.03033447265625,
-0.017547607421875,
-0.04461669921875,
-0.020172119140625,
-0.0182342529296875,
-0.073486328125,
-0.00928497314453125,
-0.0228424072265625,
-0.052520751953125,
0.0135498046875,
0.00815582275390625,
-0.0196075439453125,
0.01837158203125,
-0.037750244140625,
0.04669189453125,
-0.0286102294921875,
-0.017974853515625,
-0.007965087890625,
-0.03594970703125,
0.0226898193359375,
-0.0104827880859375,
0.0173492431640625,
-0.005313873291015625,
0.0127410888671875,
0.056549072265625,
-0.028961181640625,
0.07861328125,
0.02099609375,
-0.0250091552734375,
0.0400390625,
-0.00008124113082885742,
0.052154541015625,
0.00506591796875,
-0.016448974609375,
0.039337158203125,
-0.0111236572265625,
-0.0248870849609375,
-0.01299285888671875,
0.05731201171875,
-0.09869384765625,
-0.01074981689453125,
-0.02801513671875,
-0.03179931640625,
0.006195068359375,
-0.01169586181640625,
0.030120849609375,
0.03863525390625,
-0.0293731689453125,
0.0190582275390625,
0.033905029296875,
-0.0247802734375,
0.044586181640625,
0.0302276611328125,
-0.0233306884765625,
-0.025787353515625,
0.0673828125,
-0.006565093994140625,
0.0024433135986328125,
0.03375244140625,
-0.01517486572265625,
-0.037261962890625,
-0.0228271484375,
-0.05438232421875,
0.030029296875,
-0.039459228515625,
-0.0059051513671875,
-0.03863525390625,
-0.019012451171875,
0.004955291748046875,
0.005199432373046875,
-0.047760009765625,
-0.04449462890625,
-0.052337646484375,
-0.01258087158203125,
0.053558349609375,
0.06500244140625,
0.0235748291015625,
0.055755615234375,
-0.050079345703125,
0.0120849609375,
0.0171051025390625,
0.0166168212890625,
-0.016021728515625,
-0.052337646484375,
-0.01837158203125,
0.01082611083984375,
-0.03802490234375,
-0.049041748046875,
0.03826904296875,
0.018157958984375,
0.0247802734375,
0.0184783935546875,
-0.00283050537109375,
0.024078369140625,
-0.043731689453125,
0.07708740234375,
0.0040740966796875,
-0.05792236328125,
0.0294036865234375,
-0.0290069580078125,
-0.01526641845703125,
0.02880859375,
0.022186279296875,
-0.0202789306640625,
-0.0173187255859375,
-0.0406494140625,
-0.06036376953125,
0.0460205078125,
0.052642822265625,
0.027679443359375,
0.00562286376953125,
0.03448486328125,
-0.01265716552734375,
0.002346038818359375,
-0.080810546875,
-0.02203369140625,
-0.017303466796875,
-0.021270751953125,
-0.020751953125,
-0.0122528076171875,
-0.01983642578125,
-0.00572967529296875,
0.0706787109375,
-0.005809783935546875,
0.031463623046875,
0.016632080078125,
-0.0120697021484375,
-0.0159454345703125,
0.023345947265625,
0.05078125,
0.0282440185546875,
0.0037670135498046875,
-0.02685546875,
0.034027099609375,
-0.0235748291015625,
0.011474609375,
0.012054443359375,
-0.002719879150390625,
-0.00435638427734375,
0.0132293701171875,
0.07318115234375,
0.005672454833984375,
-0.0291900634765625,
0.032989501953125,
-0.0197296142578125,
-0.0309906005859375,
-0.047576904296875,
-0.003047943115234375,
0.0072174072265625,
0.036285400390625,
0.01534271240234375,
-0.0006551742553710938,
-0.007236480712890625,
-0.01885986328125,
0.016845703125,
0.01551055908203125,
-0.0265045166015625,
-0.022003173828125,
0.075439453125,
0.0390625,
-0.038299560546875,
0.032958984375,
-0.00998687744140625,
-0.01605224609375,
0.056671142578125,
0.03131103515625,
0.059112548828125,
-0.007419586181640625,
0.0230255126953125,
0.0267486572265625,
0.032684326171875,
-0.0158538818359375,
0.0163421630859375,
0.0211334228515625,
-0.04193115234375,
-0.0435791015625,
-0.0771484375,
-0.035491943359375,
0.0297393798828125,
-0.035736083984375,
0.03802490234375,
-0.01751708984375,
-0.0103759765625,
-0.018463134765625,
0.0026836395263671875,
-0.045166015625,
0.0364990234375,
0.00018298625946044922,
0.07965087890625,
-0.0830078125,
0.06231689453125,
0.046112060546875,
-0.061981201171875,
-0.091796875,
-0.0007534027099609375,
-0.0088653564453125,
-0.0657958984375,
0.04351806640625,
0.011474609375,
-0.01119232177734375,
0.00565338134765625,
-0.0494384765625,
-0.068603515625,
0.0806884765625,
0.0189666748046875,
-0.05731201171875,
-0.0003795623779296875,
0.0292816162109375,
0.054046630859375,
-0.04638671875,
0.0078277587890625,
0.0396728515625,
0.02105712890625,
0.0158538818359375,
-0.0743408203125,
-0.01392364501953125,
-0.0215606689453125,
-0.003955841064453125,
-0.01073455810546875,
-0.07196044921875,
0.06549072265625,
0.00815582275390625,
0.0025310516357421875,
0.03887939453125,
0.060546875,
0.00463104248046875,
0.023590087890625,
0.0257110595703125,
0.055084228515625,
0.056793212890625,
-0.00955963134765625,
0.0821533203125,
-0.040496826171875,
0.03143310546875,
0.065673828125,
-0.001678466796875,
0.080322265625,
0.0225830078125,
-0.0050201416015625,
0.069580078125,
0.061279296875,
0.01056671142578125,
0.019744873046875,
-0.0036067962646484375,
-0.00768280029296875,
-0.007274627685546875,
-0.00860595703125,
-0.039825439453125,
0.0438232421875,
0.0249176025390625,
-0.044464111328125,
0.0014095306396484375,
-0.0080108642578125,
0.01605224609375,
0.0068206787109375,
-0.01739501953125,
0.04681396484375,
0.0197296142578125,
-0.039703369140625,
0.07470703125,
0.0075836181640625,
0.03302001953125,
-0.03167724609375,
0.00045609474182128906,
-0.01715087890625,
0.0122528076171875,
-0.0161285400390625,
-0.036346435546875,
0.0149688720703125,
0.0269775390625,
-0.0018186569213867188,
-0.01160430908203125,
0.022064208984375,
-0.032379150390625,
-0.06658935546875,
0.023651123046875,
0.0333251953125,
0.037078857421875,
-0.00792694091796875,
-0.05706787109375,
0.012603759765625,
0.001605987548828125,
-0.03924560546875,
0.005176544189453125,
0.025482177734375,
-0.0141448974609375,
0.0565185546875,
0.04766845703125,
-0.01093292236328125,
-0.00943756103515625,
0.018585205078125,
0.07666015625,
-0.0194549560546875,
-0.032562255859375,
-0.034637451171875,
0.06317138671875,
0.005893707275390625,
-0.041900634765625,
0.0289306640625,
0.0281524658203125,
0.0290985107421875,
-0.004688262939453125,
0.049835205078125,
0.017822265625,
0.0296783447265625,
-0.04229736328125,
0.051177978515625,
-0.078857421875,
0.03875732421875,
-0.0240936279296875,
-0.08758544921875,
-0.017242431640625,
0.051513671875,
-0.0259246826171875,
-0.00833892822265625,
0.04833984375,
0.077880859375,
-0.00908660888671875,
-0.003307342529296875,
0.013214111328125,
0.0313720703125,
0.0249786376953125,
0.0389404296875,
0.05255126953125,
-0.05633544921875,
0.060638427734375,
-0.003208160400390625,
-0.02691650390625,
-0.019012451171875,
-0.06341552734375,
-0.0828857421875,
-0.04449462890625,
-0.01245880126953125,
-0.023834228515625,
-0.00522613525390625,
0.065673828125,
0.053192138671875,
-0.05194091796875,
-0.01296234130859375,
-0.020721435546875,
-0.01078033447265625,
-0.001178741455078125,
-0.01348876953125,
0.029876708984375,
-0.01922607421875,
-0.0672607421875,
0.04095458984375,
-0.001674652099609375,
0.0080413818359375,
-0.0166778564453125,
-0.0228118896484375,
-0.01436614990234375,
0.0257110595703125,
0.040924072265625,
0.03228759765625,
-0.0640869140625,
-0.026092529296875,
0.005527496337890625,
-0.01320648193359375,
0.007472991943359375,
-0.0005517005920410156,
-0.035308837890625,
-0.0015153884887695312,
-0.0014162063598632812,
0.039154052734375,
0.055572509765625,
0.003383636474609375,
0.033782958984375,
-0.0660400390625,
0.035614013671875,
0.0011053085327148438,
0.031158447265625,
0.02581787109375,
-0.037078857421875,
0.0421142578125,
-0.00933837890625,
-0.072265625,
-0.037567138671875,
0.0196685791015625,
-0.086669921875,
0.0013875961303710938,
0.0977783203125,
0.0012941360473632812,
-0.01515960693359375,
0.004528045654296875,
-0.04571533203125,
0.019012451171875,
-0.0384521484375,
0.05072021484375,
0.0377197265625,
0.003246307373046875,
0.001399993896484375,
-0.039459228515625,
0.042236328125,
0.03125,
-0.050445556640625,
-0.0168914794921875,
0.0247344970703125,
0.028045654296875,
0.01934814453125,
0.0309906005859375,
0.003780364990234375,
0.00778961181640625,
-0.01285552978515625,
0.02142333984375,
0.004467010498046875,
-0.044891357421875,
-0.0228118896484375,
0.0042266845703125,
0.0096588134765625,
-0.01218414306640625
]
] |
stabilityai/sd-vae-ft-ema | 2023-06-05T16:27:31.000Z | [
"diffusers",
"stable-diffusion",
"stable-diffusion-diffusers",
"license:mit",
"has_space",
"diffusers:AutoencoderKL",
"region:us"
] | null | stabilityai | null | null | stabilityai/sd-vae-ft-ema | 87 | 7,644 | diffusers | 2022-10-13T12:51:55 | ---
license: mit
tags:
- stable-diffusion
- stable-diffusion-diffusers
inference: false
---
# Improved Autoencoders
## Utilizing
These weights are intended to be used with the [🧨 diffusers library](https://github.com/huggingface/diffusers). If you are looking for the model to use with the original [CompVis Stable Diffusion codebase](https://github.com/CompVis/stable-diffusion), [come here](https://huggingface.co/stabilityai/sd-vae-ft-ema-original).
#### How to use with 🧨 diffusers
You can integrate this fine-tuned VAE decoder to your existing `diffusers` workflows, by including a `vae` argument to the `StableDiffusionPipeline`
```py
from diffusers.models import AutoencoderKL
from diffusers import StableDiffusionPipeline
model = "CompVis/stable-diffusion-v1-4"
vae = AutoencoderKL.from_pretrained("stabilityai/sd-vae-ft-ema")
pipe = StableDiffusionPipeline.from_pretrained(model, vae=vae)
```
## Decoder Finetuning
We publish two kl-f8 autoencoder versions, finetuned from the original [kl-f8 autoencoder](https://github.com/CompVis/latent-diffusion#pretrained-autoencoding-models) on a 1:1 ratio of [LAION-Aesthetics](https://laion.ai/blog/laion-aesthetics/) and LAION-Humans, an unreleased subset containing only SFW images of humans. The intent was to fine-tune on the Stable Diffusion training set (the autoencoder was originally trained on OpenImages) but also enrich the dataset with images of humans to improve the reconstruction of faces.
The first, _ft-EMA_, was resumed from the original checkpoint, trained for 313198 steps and uses EMA weights. It uses the same loss configuration as the original checkpoint (L1 + LPIPS).
The second, _ft-MSE_, was resumed from _ft-EMA_ and uses EMA weights and was trained for another 280k steps using a different loss, with more emphasis
on MSE reconstruction (MSE + 0.1 * LPIPS). It produces somewhat ``smoother'' outputs. The batch size for both versions was 192 (16 A100s, batch size 12 per GPU).
To keep compatibility with existing models, only the decoder part was finetuned; the checkpoints can be used as a drop-in replacement for the existing autoencoder.
_Original kl-f8 VAE vs f8-ft-EMA vs f8-ft-MSE_
## Evaluation
### COCO 2017 (256x256, val, 5000 images)
| Model | train steps | rFID | PSNR | SSIM | PSIM | Link | Comments
|----------|---------|------|--------------|---------------|---------------|-----------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------|
| | | | | | | | |
| original | 246803 | 4.99 | 23.4 +/- 3.8 | 0.69 +/- 0.14 | 1.01 +/- 0.28 | https://ommer-lab.com/files/latent-diffusion/kl-f8.zip | as used in SD |
| ft-EMA | 560001 | 4.42 | 23.8 +/- 3.9 | 0.69 +/- 0.13 | 0.96 +/- 0.27 | https://huggingface.co/stabilityai/sd-vae-ft-ema-original/resolve/main/vae-ft-ema-560000-ema-pruned.ckpt | slightly better overall, with EMA |
| ft-MSE | 840001 | 4.70 | 24.5 +/- 3.7 | 0.71 +/- 0.13 | 0.92 +/- 0.27 | https://huggingface.co/stabilityai/sd-vae-ft-mse-original/resolve/main/vae-ft-mse-840000-ema-pruned.ckpt | resumed with EMA from ft-EMA, emphasis on MSE (rec. loss = MSE + 0.1 * LPIPS), smoother outputs |
### LAION-Aesthetics 5+ (256x256, subset, 10000 images)
| Model | train steps | rFID | PSNR | SSIM | PSIM | Link | Comments
|----------|-----------|------|--------------|---------------|---------------|-----------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------|
| | | | | | | | |
| original | 246803 | 2.61 | 26.0 +/- 4.4 | 0.81 +/- 0.12 | 0.75 +/- 0.36 | https://ommer-lab.com/files/latent-diffusion/kl-f8.zip | as used in SD |
| ft-EMA | 560001 | 1.77 | 26.7 +/- 4.8 | 0.82 +/- 0.12 | 0.67 +/- 0.34 | https://huggingface.co/stabilityai/sd-vae-ft-ema-original/resolve/main/vae-ft-ema-560000-ema-pruned.ckpt | slightly better overall, with EMA |
| ft-MSE | 840001 | 1.88 | 27.3 +/- 4.7 | 0.83 +/- 0.11 | 0.65 +/- 0.34 | https://huggingface.co/stabilityai/sd-vae-ft-mse-original/resolve/main/vae-ft-mse-840000-ema-pruned.ckpt | resumed with EMA from ft-EMA, emphasis on MSE (rec. loss = MSE + 0.1 * LPIPS), smoother outputs |
### Visual
_Visualization of reconstructions on 256x256 images from the COCO2017 validation dataset._
<p align="center">
<br>
<b>
256x256: ft-EMA (left), ft-MSE (middle), original (right)</b>
</p>
<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00025_merged.png />
</p>
<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00011_merged.png />
</p>
<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00037_merged.png />
</p>
<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00043_merged.png />
</p>
<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00053_merged.png />
</p>
<p align="center">
<img src=https://huggingface.co/stabilityai/stable-diffusion-decoder-finetune/resolve/main/eval/ae-decoder-tuning-reconstructions/merged/00029_merged.png />
</p>
| 6,839 | [
[
-0.054412841796875,
-0.0312042236328125,
0.01244354248046875,
0.0175018310546875,
-0.01038360595703125,
-0.0122222900390625,
-0.0011138916015625,
-0.0038089752197265625,
0.040435791015625,
0.0182647705078125,
-0.025604248046875,
-0.032684326171875,
-0.045166015625,
0.01081085205078125,
-0.0025234222412109375,
0.04803466796875,
-0.00679779052734375,
0.01136016845703125,
0.006710052490234375,
-0.01947021484375,
-0.032562255859375,
-0.023895263671875,
-0.05474853515625,
-0.0155487060546875,
0.0310516357421875,
0.0110626220703125,
0.026885986328125,
0.04425048828125,
0.0232086181640625,
0.0214996337890625,
-0.01959228515625,
-0.00022983551025390625,
-0.027984619140625,
-0.0184478759765625,
0.01910400390625,
-0.011932373046875,
-0.051055908203125,
-0.0089874267578125,
0.045440673828125,
0.02764892578125,
-0.01258087158203125,
0.00695037841796875,
0.01873779296875,
0.059967041015625,
-0.046966552734375,
0.006832122802734375,
-0.00920867919921875,
0.005771636962890625,
-0.0038051605224609375,
-0.0287628173828125,
-0.01526641845703125,
-0.01425933837890625,
-0.01053619384765625,
-0.0555419921875,
0.0173797607421875,
-0.0021915435791015625,
0.11419677734375,
0.022613525390625,
-0.0233306884765625,
-0.004627227783203125,
-0.03997802734375,
0.049224853515625,
-0.057891845703125,
0.033843994140625,
0.0175933837890625,
0.012939453125,
-0.00811767578125,
-0.045440673828125,
-0.033416748046875,
0.0050201416015625,
-0.026702880859375,
0.034881591796875,
-0.0179901123046875,
-0.0008349418640136719,
0.0330810546875,
0.04425048828125,
-0.0465087890625,
-0.006649017333984375,
-0.044830322265625,
-0.0248565673828125,
0.044921875,
0.01091766357421875,
0.0175018310546875,
-0.0211334228515625,
-0.01922607421875,
-0.03411865234375,
-0.0408935546875,
0.01273345947265625,
0.017364501953125,
-0.01268768310546875,
-0.0313720703125,
0.0259857177734375,
-0.0188751220703125,
0.044403076171875,
0.01204681396484375,
-0.012176513671875,
0.04400634765625,
-0.0287017822265625,
-0.040496826171875,
-0.0084381103515625,
0.07342529296875,
0.03216552734375,
-0.01035308837890625,
0.018402099609375,
-0.01165771484375,
0.003589630126953125,
-0.004718780517578125,
-0.0772705078125,
-0.029052734375,
0.0258941650390625,
-0.05419921875,
-0.0251617431640625,
0.005428314208984375,
-0.0767822265625,
0.0198211669921875,
-0.01515960693359375,
0.031585693359375,
-0.04052734375,
-0.0238037109375,
0.0040130615234375,
-0.019195556640625,
0.050994873046875,
0.034820556640625,
-0.053375244140625,
0.0219879150390625,
0.00859832763671875,
0.0633544921875,
-0.00072479248046875,
0.00435638427734375,
-0.02545166015625,
-0.0040130615234375,
-0.034210205078125,
0.04150390625,
-0.00870513916015625,
-0.0285491943359375,
-0.0240325927734375,
0.02838134765625,
-0.01459503173828125,
-0.0289154052734375,
0.05792236328125,
-0.033416748046875,
0.0135040283203125,
-0.027374267578125,
-0.031494140625,
-0.0201263427734375,
0.00994873046875,
-0.054840087890625,
0.081298828125,
0.01149749755859375,
-0.063232421875,
0.0217742919921875,
-0.046478271484375,
-0.00725555419921875,
-0.0272064208984375,
-0.0149078369140625,
-0.05059814453125,
-0.00997161865234375,
0.041046142578125,
0.0259246826171875,
-0.018035888671875,
0.00004547834396362305,
-0.0174407958984375,
-0.021820068359375,
0.0018167495727539062,
-0.048065185546875,
0.0902099609375,
0.031097412109375,
-0.035919189453125,
-0.0011892318725585938,
-0.0694580078125,
-0.00547027587890625,
0.0274200439453125,
-0.025299072265625,
-0.0009188652038574219,
-0.02630615234375,
-0.003955841064453125,
0.0304412841796875,
0.0016546249389648438,
-0.027740478515625,
0.004039764404296875,
-0.031982421875,
0.03466796875,
0.06268310546875,
0.0165863037109375,
0.032562255859375,
-0.03778076171875,
0.035614013671875,
0.021759033203125,
0.005748748779296875,
-0.0238037109375,
-0.04779052734375,
-0.0765380859375,
-0.0352783203125,
0.0308990478515625,
0.036834716796875,
-0.03570556640625,
0.044342041015625,
-0.0247039794921875,
-0.035858154296875,
-0.0478515625,
-0.003459930419921875,
0.0101776123046875,
0.030120849609375,
0.0325927734375,
-0.035675048828125,
-0.042755126953125,
-0.06536865234375,
0.01491546630859375,
0.00884246826171875,
-0.0005974769592285156,
0.027740478515625,
0.048583984375,
0.003978729248046875,
0.064697265625,
-0.051605224609375,
-0.0302734375,
0.015869140625,
-0.004245758056640625,
0.0283050537109375,
0.060546875,
0.0662841796875,
-0.053436279296875,
-0.055877685546875,
-0.0084686279296875,
-0.0634765625,
0.0006837844848632812,
-0.0008425712585449219,
-0.0229644775390625,
0.01062774658203125,
0.0302276611328125,
-0.03924560546875,
0.05670166015625,
0.0325927734375,
-0.02423095703125,
0.037567138671875,
-0.0277557373046875,
0.02880859375,
-0.08734130859375,
0.02838134765625,
0.00638580322265625,
-0.0262298583984375,
-0.030242919921875,
-0.0089263916015625,
0.0006918907165527344,
-0.00174713134765625,
-0.035552978515625,
0.04083251953125,
-0.057373046875,
0.0017900466918945312,
0.005283355712890625,
-0.002262115478515625,
0.0168914794921875,
0.057037353515625,
0.002452850341796875,
0.04766845703125,
0.052703857421875,
-0.0369873046875,
0.0050048828125,
0.0129241943359375,
-0.020172119140625,
0.031707763671875,
-0.06329345703125,
0.0024394989013671875,
-0.02288818359375,
0.0294342041015625,
-0.0775146484375,
-0.01453399658203125,
0.048797607421875,
-0.04241943359375,
0.043701171875,
-0.0231170654296875,
-0.0214996337890625,
-0.03564453125,
-0.0311431884765625,
0.0276031494140625,
0.05340576171875,
-0.03143310546875,
0.045257568359375,
0.00992584228515625,
0.021270751953125,
-0.03955078125,
-0.0535888671875,
-0.0182952880859375,
-0.01849365234375,
-0.04620361328125,
0.031463623046875,
-0.0162353515625,
0.0095672607421875,
0.0076141357421875,
-0.0045013427734375,
0.004840850830078125,
-0.007904052734375,
0.034423828125,
0.023345947265625,
-0.0262451171875,
-0.03436279296875,
0.01108551025390625,
-0.01216888427734375,
-0.003360748291015625,
-0.00844573974609375,
0.0458984375,
-0.0079498291015625,
-0.019989013671875,
-0.0533447265625,
0.0137939453125,
0.055877685546875,
-0.0200958251953125,
0.060943603515625,
0.06378173828125,
-0.029693603515625,
0.01153564453125,
-0.033294677734375,
-0.01474761962890625,
-0.037567138671875,
-0.0007500648498535156,
-0.037078857421875,
-0.05029296875,
0.056488037109375,
0.01061248779296875,
0.0032596588134765625,
0.07244873046875,
0.035064697265625,
-0.01528167724609375,
0.08306884765625,
0.0115814208984375,
0.001247406005859375,
0.033416748046875,
-0.075439453125,
0.0023365020751953125,
-0.07562255859375,
-0.0233306884765625,
-0.03887939453125,
-0.0169525146484375,
-0.0297088623046875,
-0.050201416015625,
0.029815673828125,
0.0251007080078125,
-0.01934814453125,
0.0190277099609375,
-0.0557861328125,
0.01172637939453125,
0.0221710205078125,
0.0171966552734375,
0.002902984619140625,
0.0198211669921875,
-0.018890380859375,
-0.0023708343505859375,
-0.06036376953125,
-0.034149169921875,
0.080078125,
0.0322265625,
0.0496826171875,
0.0036296844482421875,
0.0491943359375,
0.01654052734375,
0.023834228515625,
-0.034149169921875,
0.029937744140625,
-0.00472259521484375,
-0.043182373046875,
0.0002963542938232422,
-0.022369384765625,
-0.07330322265625,
0.028350830078125,
-0.0166015625,
-0.056396484375,
0.048004150390625,
0.0325927734375,
-0.0245361328125,
0.035003662109375,
-0.0533447265625,
0.07708740234375,
-0.005840301513671875,
-0.028594970703125,
0.003368377685546875,
-0.046142578125,
0.018463134765625,
0.0235595703125,
0.01235198974609375,
-0.004055023193359375,
0.004726409912109375,
0.07037353515625,
-0.05755615234375,
0.05419921875,
-0.01280975341796875,
-0.0091552734375,
0.047637939453125,
-0.00864410400390625,
0.039031982421875,
0.003742218017578125,
-0.002269744873046875,
0.0225067138671875,
0.005100250244140625,
-0.03948974609375,
-0.0296173095703125,
0.06591796875,
-0.07086181640625,
-0.034576416015625,
-0.0477294921875,
-0.006927490234375,
0.034454345703125,
0.019195556640625,
0.048004150390625,
0.049713134765625,
-0.0068817138671875,
0.016876220703125,
0.0577392578125,
-0.0180511474609375,
0.0394287109375,
0.018524169921875,
-0.005069732666015625,
-0.057159423828125,
0.0828857421875,
0.01494598388671875,
0.022247314453125,
0.0264739990234375,
0.005718231201171875,
-0.01044464111328125,
-0.032012939453125,
-0.038665771484375,
0.022979736328125,
-0.056396484375,
-0.0296630859375,
-0.06640625,
-0.028228759765625,
-0.038055419921875,
-0.01605224609375,
-0.03607177734375,
-0.0290679931640625,
-0.050140380859375,
0.01140594482421875,
0.02777099609375,
0.0306854248046875,
-0.0181732177734375,
0.0120086669921875,
-0.05633544921875,
0.0179443359375,
0.004367828369140625,
0.017730712890625,
0.01251220703125,
-0.044677734375,
-0.009246826171875,
0.0108642578125,
-0.041046142578125,
-0.0775146484375,
0.051025390625,
0.009246826171875,
0.049163818359375,
0.027252197265625,
-0.01090240478515625,
0.059326171875,
-0.0164031982421875,
0.058013916015625,
0.0157318115234375,
-0.051483154296875,
0.04974365234375,
-0.019989013671875,
0.0149078369140625,
0.0283355712890625,
0.03814697265625,
-0.0175628662109375,
-0.005100250244140625,
-0.05572509765625,
-0.0697021484375,
0.0555419921875,
0.03814697265625,
-0.025146484375,
0.006275177001953125,
0.0243377685546875,
-0.01218414306640625,
-0.0011310577392578125,
-0.052276611328125,
-0.054656982421875,
-0.03558349609375,
-0.0103912353515625,
-0.0003657341003417969,
-0.004756927490234375,
-0.004604339599609375,
-0.041717529296875,
0.058319091796875,
-0.003482818603515625,
0.048736572265625,
0.03814697265625,
-0.01042938232421875,
-0.0113525390625,
0.005390167236328125,
0.035919189453125,
0.030426025390625,
-0.036651611328125,
-0.00472259521484375,
0.01480865478515625,
-0.0288848876953125,
0.01641845703125,
-0.00251007080078125,
-0.034210205078125,
0.0019102096557617188,
0.01334381103515625,
0.0689697265625,
-0.00556182861328125,
-0.0193328857421875,
0.041473388671875,
-0.0164642333984375,
-0.0318603515625,
-0.03460693359375,
0.00838470458984375,
0.006298065185546875,
-0.0030422210693359375,
0.0233917236328125,
0.029815673828125,
0.01404571533203125,
-0.026275634765625,
0.01526641845703125,
0.0228118896484375,
-0.0205230712890625,
-0.0243682861328125,
0.06494140625,
0.0001494884490966797,
-0.0038814544677734375,
0.0333251953125,
-0.02667236328125,
-0.0361328125,
0.06658935546875,
0.03350830078125,
0.064697265625,
-0.0203704833984375,
0.0012540817260742188,
0.07879638671875,
0.01558685302734375,
-0.0031642913818359375,
0.0209808349609375,
-0.0020008087158203125,
-0.03558349609375,
-0.0147857666015625,
-0.06158447265625,
0.023834228515625,
0.0124969482421875,
-0.057281494140625,
0.031219482421875,
-0.03131103515625,
-0.0188140869140625,
0.002964019775390625,
0.0027828216552734375,
-0.06817626953125,
0.0286712646484375,
-0.0033245086669921875,
0.07366943359375,
-0.0743408203125,
0.06085205078125,
0.042205810546875,
-0.04736328125,
-0.0689697265625,
-0.006366729736328125,
-0.00820159912109375,
-0.0325927734375,
0.037628173828125,
0.00882720947265625,
0.007366180419921875,
0.0100250244140625,
-0.031219482421875,
-0.07269287109375,
0.1119384765625,
0.0190277099609375,
-0.043304443359375,
0.0016994476318359375,
-0.0192413330078125,
0.034515380859375,
-0.034454345703125,
0.046478271484375,
0.0357666015625,
0.0288238525390625,
0.02044677734375,
-0.050323486328125,
0.02496337890625,
-0.0308074951171875,
0.01715087890625,
0.01287078857421875,
-0.07415771484375,
0.059967041015625,
-0.00849151611328125,
-0.0169525146484375,
0.0012836456298828125,
0.0638427734375,
0.02020263671875,
0.01247406005859375,
0.0401611328125,
0.06842041015625,
0.033966064453125,
-0.0173187255859375,
0.07525634765625,
-0.024169921875,
0.041778564453125,
0.05816650390625,
0.027618408203125,
0.04656982421875,
0.0333251953125,
-0.0246429443359375,
0.033538818359375,
0.06591796875,
-0.01204681396484375,
0.04156494140625,
0.00591278076171875,
-0.0022106170654296875,
-0.0047149658203125,
0.007602691650390625,
-0.044830322265625,
0.01209259033203125,
0.032470703125,
-0.042022705078125,
-0.0036983489990234375,
-0.002471923828125,
0.0163421630859375,
-0.02996826171875,
-0.0132904052734375,
0.036651611328125,
0.005176544189453125,
-0.04522705078125,
0.07806396484375,
-0.0005488395690917969,
0.05316162109375,
-0.035858154296875,
-0.015777587890625,
-0.01160430908203125,
0.00478363037109375,
-0.0301666259765625,
-0.053955078125,
0.0292816162109375,
-0.0196380615234375,
0.00018787384033203125,
-0.0054779052734375,
0.050994873046875,
-0.0262603759765625,
-0.036651611328125,
0.034759521484375,
0.00870513916015625,
0.0218048095703125,
0.0214080810546875,
-0.06585693359375,
0.0302734375,
0.016082763671875,
-0.03900146484375,
0.0244140625,
0.01338958740234375,
0.0169219970703125,
0.02764892578125,
0.0478515625,
0.005245208740234375,
0.0263519287109375,
-0.00618743896484375,
0.08221435546875,
-0.04522705078125,
-0.035064697265625,
-0.042877197265625,
0.052215576171875,
-0.0179443359375,
-0.0245361328125,
0.06427001953125,
0.05718994140625,
0.05511474609375,
-0.008026123046875,
0.04180908203125,
-0.033599853515625,
0.005031585693359375,
-0.039703369140625,
0.048004150390625,
-0.0697021484375,
0.0184478759765625,
-0.027679443359375,
-0.0626220703125,
-0.006145477294921875,
0.059661865234375,
-0.00893402099609375,
0.027252197265625,
0.045257568359375,
0.0811767578125,
-0.017120361328125,
-0.0211334228515625,
0.0050811767578125,
0.0233154296875,
0.034210205078125,
0.048736572265625,
0.0240020751953125,
-0.08282470703125,
0.041717529296875,
-0.048980712890625,
-0.01050567626953125,
-0.0241241455078125,
-0.05303955078125,
-0.055908203125,
-0.058929443359375,
-0.049896240234375,
-0.056640625,
-0.007137298583984375,
0.056182861328125,
0.0693359375,
-0.056427001953125,
-0.0115966796875,
-0.017578125,
0.0006022453308105469,
-0.02362060546875,
-0.020965576171875,
0.04473876953125,
0.00905609130859375,
-0.05963134765625,
0.012115478515625,
0.01447296142578125,
0.030303955078125,
-0.0097808837890625,
-0.0389404296875,
-0.0294036865234375,
0.003368377685546875,
0.0294342041015625,
0.0399169921875,
-0.04730224609375,
-0.0034236907958984375,
-0.0115203857421875,
-0.01230621337890625,
0.03668212890625,
0.038543701171875,
-0.04833984375,
0.0401611328125,
0.0657958984375,
0.0085906982421875,
0.06524658203125,
0.0002765655517578125,
0.0239715576171875,
-0.033416748046875,
0.011962890625,
0.002323150634765625,
0.035675048828125,
0.01509857177734375,
-0.027679443359375,
0.038787841796875,
0.036346435546875,
-0.045318603515625,
-0.05303955078125,
-0.0265960693359375,
-0.09600830078125,
-0.01020050048828125,
0.07269287109375,
-0.020050048828125,
-0.052215576171875,
0.0067901611328125,
-0.0230712890625,
0.032684326171875,
-0.052520751953125,
0.017425537109375,
0.0355224609375,
-0.0030727386474609375,
-0.0267181396484375,
-0.048248291015625,
0.04498291015625,
0.03033447265625,
-0.051361083984375,
-0.0192718505859375,
0.030914306640625,
0.044158935546875,
0.023712158203125,
0.0723876953125,
-0.0293731689453125,
0.01435089111328125,
0.01082611083984375,
0.0030918121337890625,
-0.0004870891571044922,
-0.0084991455078125,
-0.03826904296875,
0.0115966796875,
-0.013275146484375,
-0.0303955078125
]
] |
nlpie/tiny-clinicalbert | 2023-05-26T14:24:31.000Z | [
"transformers",
"pytorch",
"bert",
"fill-mask",
"arxiv:2302.04725",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | fill-mask | nlpie | null | null | nlpie/tiny-clinicalbert | 0 | 7,639 | transformers | 2023-02-10T10:36:05 | ---
title: README
emoji: 🏃
colorFrom: gray
colorTo: purple
sdk: static
pinned: false
license: mit
---
# Model Description
TinyClinicalBERT is a distilled version of the [BioClinicalBERT](https://huggingface.co/emilyalsentzer/Bio_ClinicalBERT) which is distilled for 3 epochs using a total batch size of 192 on the MIMIC-III notes dataset.
# Distillation Procedure
This model uses a unique distillation method called ‘transformer-layer distillation’ which is applied on each layer of the student to align the attention maps and the hidden states of the student with those of the teacher.
# Architecture and Initialisation
This model uses 4 hidden layers with a hidden dimension size and an embedding size of 768 resulting in a total of 15M parameters. Due to the model's small hidden dimension size, it uses random initialisation.
# Citation
If you use this model, please consider citing the following paper:
```bibtex
@misc{https://doi.org/10.48550/arxiv.2302.04725,
doi = {10.48550/ARXIV.2302.04725},
url = {https://arxiv.org/abs/2302.04725},
author = {Rohanian, Omid and Nouriborji, Mohammadmahdi and Jauncey, Hannah and Kouchaki, Samaneh and Group, ISARIC Clinical Characterisation and Clifton, Lei and Merson, Laura and Clifton, David A.},
keywords = {Computation and Language (cs.CL), Artificial Intelligence (cs.AI), Machine Learning (cs.LG), FOS: Computer and information sciences, FOS: Computer and information sciences, I.2.7, 68T50},
title = {Lightweight Transformers for Clinical Natural Language Processing},
publisher = {arXiv},
year = {2023},
copyright = {arXiv.org perpetual, non-exclusive license}
}
``` | 1,641 | [
[
-0.00989532470703125,
-0.047943115234375,
0.05816650390625,
0.004573822021484375,
-0.00879669189453125,
0.00019228458404541016,
-0.025390625,
-0.0225067138671875,
0.009429931640625,
0.031341552734375,
-0.027618408203125,
-0.042938232421875,
-0.04449462890625,
0.0070037841796875,
-0.01160430908203125,
0.09149169921875,
0.008056640625,
0.04156494140625,
-0.01885986328125,
-0.03240966796875,
-0.00844573974609375,
-0.0193023681640625,
-0.05462646484375,
-0.039154052734375,
0.03369140625,
0.032135009765625,
0.046295166015625,
0.0535888671875,
0.0390625,
0.022308349609375,
-0.02532958984375,
-0.006420135498046875,
-0.040252685546875,
-0.0266265869140625,
-0.005340576171875,
-0.04888916015625,
-0.035858154296875,
0.007045745849609375,
0.0287933349609375,
0.071533203125,
-0.015838623046875,
0.039581298828125,
0.017791748046875,
0.052581787109375,
-0.03765869140625,
0.005809783935546875,
-0.0294036865234375,
0.0015668869018554688,
-0.0007119178771972656,
0.0183563232421875,
-0.040557861328125,
-0.03497314453125,
0.033203125,
-0.0157318115234375,
0.052825927734375,
0.0025043487548828125,
0.04107666015625,
0.0245361328125,
-0.021820068359375,
-0.0307464599609375,
-0.052398681640625,
0.057464599609375,
-0.04425048828125,
-0.01110076904296875,
0.01837158203125,
0.04486083984375,
0.0103759765625,
-0.102783203125,
-0.036376953125,
-0.03155517578125,
-0.0242767333984375,
-0.00821685791015625,
-0.02252197265625,
0.01136016845703125,
0.0280914306640625,
0.0232086181640625,
-0.054656982421875,
-0.01031494140625,
-0.06982421875,
-0.0140228271484375,
0.0235595703125,
0.0228729248046875,
0.0010166168212890625,
-0.0157012939453125,
-0.040252685546875,
-0.002712249755859375,
-0.01806640625,
0.005954742431640625,
0.00240325927734375,
0.0250396728515625,
-0.0223236083984375,
0.03045654296875,
0.0089874267578125,
0.03509521484375,
0.0177001953125,
-0.024932861328125,
0.0312347412109375,
-0.0249786376953125,
-0.022674560546875,
0.037872314453125,
0.080810546875,
-0.0009074211120605469,
0.020050048828125,
0.0211181640625,
0.0178985595703125,
-0.00339508056640625,
0.030975341796875,
-0.08197021484375,
-0.05780029296875,
0.0352783203125,
-0.044830322265625,
-0.043548583984375,
-0.0176849365234375,
-0.032379150390625,
-0.0247344970703125,
-0.0164794921875,
0.0211029052734375,
-0.04364013671875,
0.01203155517578125,
0.004322052001953125,
0.006679534912109375,
-0.027252197265625,
0.0177154541015625,
-0.05780029296875,
0.036865234375,
0.0023441314697265625,
0.066650390625,
-0.0053863525390625,
0.01050567626953125,
-0.01812744140625,
0.00809478759765625,
0.00689697265625,
0.034393310546875,
-0.01263427734375,
-0.033294677734375,
-0.0172119140625,
0.0239105224609375,
-0.005748748779296875,
-0.05572509765625,
0.048187255859375,
-0.00795745849609375,
0.0117950439453125,
0.0122528076171875,
0.0017910003662109375,
-0.01476287841796875,
0.0030517578125,
-0.060394287109375,
0.0706787109375,
0.00860595703125,
-0.0614013671875,
0.01806640625,
-0.033660888671875,
-0.0108642578125,
0.0058135986328125,
-0.01038360595703125,
-0.048797607421875,
0.0114288330078125,
0.0158233642578125,
0.019317626953125,
-0.026580810546875,
0.0258636474609375,
-0.032470703125,
-0.0221405029296875,
0.0142822265625,
-0.02862548828125,
0.0802001953125,
0.011962890625,
-0.0173797607421875,
0.001483917236328125,
-0.09014892578125,
0.01165008544921875,
0.00557708740234375,
0.0009531974792480469,
-0.031524658203125,
-0.030853271484375,
0.01611328125,
0.02044677734375,
0.0302276611328125,
-0.04364013671875,
0.019927978515625,
-0.0249786376953125,
0.050018310546875,
0.0215301513671875,
0.0118560791015625,
0.02117919921875,
-0.0303802490234375,
0.0299835205078125,
0.0156097412109375,
0.030609130859375,
0.032501220703125,
-0.0274658203125,
-0.049102783203125,
-0.06329345703125,
0.042022705078125,
0.032958984375,
-0.047119140625,
0.0042266845703125,
-0.00290679931640625,
-0.053253173828125,
-0.07586669921875,
0.01375579833984375,
0.03594970703125,
0.04296875,
0.043121337890625,
0.00005334615707397461,
-0.027557373046875,
-0.09906005859375,
0.00841522216796875,
-0.014404296875,
-0.0092926025390625,
0.0292816162109375,
0.04669189453125,
-0.044403076171875,
0.0537109375,
-0.055267333984375,
-0.053802490234375,
-0.0258941650390625,
0.036651611328125,
0.04010009765625,
0.037017822265625,
0.0577392578125,
-0.039886474609375,
-0.052093505859375,
-0.0194549560546875,
-0.06317138671875,
-0.007762908935546875,
-0.0040435791015625,
0.009185791015625,
-0.0081787109375,
0.0215911865234375,
-0.027984619140625,
0.042205810546875,
0.024688720703125,
0.019012451171875,
0.01258087158203125,
-0.05072021484375,
-0.0002899169921875,
-0.07843017578125,
0.0245361328125,
0.004802703857421875,
-0.0292205810546875,
-0.039154052734375,
-0.005878448486328125,
0.01314544677734375,
-0.00933074951171875,
-0.0579833984375,
0.024017333984375,
-0.040863037109375,
0.04388427734375,
-0.0253753662109375,
0.0228729248046875,
0.0086212158203125,
0.037078857421875,
-0.013336181640625,
0.03741455078125,
0.0286712646484375,
-0.0178985595703125,
0.0238037109375,
0.0189056396484375,
-0.003108978271484375,
0.0226593017578125,
-0.08544921875,
0.00832366943359375,
-0.00516510009765625,
0.01544189453125,
-0.0863037109375,
-0.004177093505859375,
0.0041046142578125,
-0.028411865234375,
0.023773193359375,
-0.011444091796875,
-0.05133056640625,
-0.038848876953125,
-0.040557861328125,
0.047119140625,
0.04168701171875,
-0.023193359375,
0.037872314453125,
0.006938934326171875,
-0.0007767677307128906,
-0.013092041015625,
-0.06658935546875,
-0.02911376953125,
-0.0150146484375,
-0.049041748046875,
0.022857666015625,
-0.01611328125,
-0.0170745849609375,
0.0004355907440185547,
0.01055145263671875,
-0.012664794921875,
-0.0157318115234375,
0.0293426513671875,
0.034759521484375,
-0.01026153564453125,
-0.0029144287109375,
0.03594970703125,
-0.015472412109375,
0.005947113037109375,
-0.002346038818359375,
0.0228729248046875,
-0.0280914306640625,
-0.004146575927734375,
-0.06951904296875,
0.010833740234375,
0.052215576171875,
0.0284576416015625,
0.07080078125,
0.056121826171875,
-0.0328369140625,
-0.002651214599609375,
-0.0236358642578125,
-0.03076171875,
-0.038177490234375,
0.0102691650390625,
-0.0157623291015625,
-0.048492431640625,
0.052215576171875,
-0.0037021636962890625,
-0.0223388671875,
0.048858642578125,
0.06964111328125,
-0.0098114013671875,
0.08343505859375,
0.0443115234375,
0.015289306640625,
0.0034961700439453125,
-0.041229248046875,
0.005584716796875,
-0.06231689453125,
-0.014801025390625,
-0.040557861328125,
-0.04107666015625,
-0.04949951171875,
-0.0233001708984375,
0.06329345703125,
0.0080718994140625,
-0.01519775390625,
0.0216064453125,
-0.06683349609375,
0.00984954833984375,
0.036163330078125,
0.034759521484375,
0.0278778076171875,
-0.00959014892578125,
-0.04754638671875,
-0.0272064208984375,
-0.050750732421875,
-0.051483154296875,
0.0894775390625,
0.0228729248046875,
0.05804443359375,
0.01253509521484375,
0.0643310546875,
0.0241241455078125,
0.036590576171875,
-0.03814697265625,
0.0285491943359375,
-0.0057220458984375,
-0.07080078125,
0.0015077590942382812,
-0.0136871337890625,
-0.0606689453125,
0.0191192626953125,
-0.01561737060546875,
-0.0721435546875,
0.0178985595703125,
0.0180511474609375,
-0.057464599609375,
0.0017099380493164062,
-0.0594482421875,
0.056976318359375,
-0.00020837783813476562,
-0.048858642578125,
-0.005035400390625,
-0.07275390625,
0.0404052734375,
-0.0192108154296875,
0.0178985595703125,
0.01593017578125,
0.01132965087890625,
0.044097900390625,
-0.05023193359375,
0.0767822265625,
-0.00603485107421875,
0.024932861328125,
0.04034423828125,
0.005199432373046875,
0.020172119140625,
0.03729248046875,
0.00826263427734375,
0.0207366943359375,
0.0523681640625,
-0.028900146484375,
-0.0258331298828125,
0.06488037109375,
-0.06402587890625,
-0.035614013671875,
-0.0714111328125,
-0.0255584716796875,
0.0014562606811523438,
0.0154876708984375,
0.025360107421875,
0.037933349609375,
-0.01629638671875,
0.031524658203125,
0.0450439453125,
-0.01080322265625,
0.0156402587890625,
0.04107666015625,
-0.002574920654296875,
-0.035858154296875,
0.036773681640625,
-0.0005540847778320312,
0.0288238525390625,
0.00418853759765625,
0.02545166015625,
-0.017547607421875,
-0.0357666015625,
-0.029083251953125,
0.0428466796875,
-0.037017822265625,
0.007465362548828125,
-0.06219482421875,
-0.0292816162109375,
-0.0281524658203125,
-0.0007295608520507812,
-0.04010009765625,
-0.03558349609375,
-0.059234619140625,
0.0173492431640625,
0.0227203369140625,
0.027557373046875,
-0.0157623291015625,
0.0227508544921875,
-0.06549072265625,
0.022247314453125,
0.004924774169921875,
-0.0152740478515625,
0.0011262893676757812,
-0.05450439453125,
-0.021820068359375,
0.01139068603515625,
-0.040618896484375,
-0.03924560546875,
0.045562744140625,
0.01047515869140625,
0.0318603515625,
0.03411865234375,
0.0239715576171875,
0.045013427734375,
-0.047637939453125,
0.064208984375,
0.0204925537109375,
-0.0411376953125,
0.052398681640625,
-0.00925445556640625,
0.028350830078125,
0.09088134765625,
0.048065185546875,
0.00475311279296875,
-0.0249786376953125,
-0.0699462890625,
-0.0609130859375,
0.05096435546875,
0.0241851806640625,
-0.01335906982421875,
0.01486968994140625,
0.0255584716796875,
0.0170745849609375,
0.0248565673828125,
-0.053619384765625,
-0.03564453125,
0.01666259765625,
-0.0433349609375,
0.0058441162109375,
-0.01971435546875,
-0.045318603515625,
-0.047119140625,
0.03717041015625,
-0.0165252685546875,
0.0185394287109375,
0.01776123046875,
-0.00719451904296875,
0.00821685791015625,
-0.0004949569702148438,
0.05914306640625,
0.042633056640625,
-0.048126220703125,
0.0110015869140625,
0.00914764404296875,
-0.0650634765625,
0.0020847320556640625,
0.0100250244140625,
0.00888824462890625,
0.01050567626953125,
0.034149169921875,
0.033477783203125,
0.01308441162109375,
-0.041015625,
0.0404052734375,
-0.0246429443359375,
-0.034912109375,
-0.049835205078125,
0.0159912109375,
-0.001041412353515625,
0.040985107421875,
0.015350341796875,
0.017669677734375,
0.03497314453125,
-0.039581298828125,
0.0047454833984375,
0.0004055500030517578,
-0.05462646484375,
-0.0267181396484375,
0.08599853515625,
-0.009918212890625,
-0.007801055908203125,
0.048980712890625,
-0.0002257823944091797,
-0.0025997161865234375,
0.031280517578125,
0.021942138671875,
0.065673828125,
0.01213836669921875,
0.0030307769775390625,
0.034576416015625,
0.008636474609375,
-0.0004520416259765625,
0.0163116455078125,
-0.0036163330078125,
-0.01381683349609375,
0.0162353515625,
-0.051727294921875,
-0.0210418701171875,
0.007671356201171875,
-0.07183837890625,
0.023468017578125,
-0.053009033203125,
-0.032989501953125,
0.03131103515625,
0.01165771484375,
-0.04608154296875,
0.0110015869140625,
0.0008778572082519531,
0.056304931640625,
-0.0592041015625,
0.08209228515625,
0.046173095703125,
-0.0469970703125,
-0.060394287109375,
-0.0138397216796875,
0.019500732421875,
-0.049224853515625,
0.0643310546875,
-0.00656890869140625,
0.0003132820129394531,
-0.0133056640625,
-0.03936767578125,
-0.0374755859375,
0.09942626953125,
0.0132904052734375,
-0.05078125,
-0.005870819091796875,
0.01012420654296875,
0.07830810546875,
-0.0271453857421875,
0.034423828125,
0.04742431640625,
0.003986358642578125,
0.0302886962890625,
-0.071533203125,
0.00014674663543701172,
-0.0195465087890625,
0.0217742919921875,
-0.0160064697265625,
-0.058563232421875,
0.07080078125,
-0.034332275390625,
0.00553131103515625,
0.02825927734375,
0.033660888671875,
0.0404052734375,
0.035064697265625,
0.0140380859375,
0.033416748046875,
0.044647216796875,
-0.00531005859375,
0.066162109375,
-0.04095458984375,
0.044952392578125,
0.084716796875,
-0.01861572265625,
0.040985107421875,
0.051116943359375,
-0.0165863037109375,
0.05157470703125,
0.03485107421875,
-0.0013151168823242188,
0.0633544921875,
0.005901336669921875,
-0.0228271484375,
0.0063323974609375,
0.0074005126953125,
-0.035797119140625,
0.0167083740234375,
0.0181884765625,
-0.0726318359375,
-0.006343841552734375,
0.022216796875,
0.007114410400390625,
-0.0014019012451171875,
-0.00426483154296875,
0.04315185546875,
0.0193023681640625,
-0.0218963623046875,
0.0753173828125,
0.00021183490753173828,
0.0154876708984375,
-0.0394287109375,
0.00402069091796875,
0.001056671142578125,
0.056915283203125,
-0.00873565673828125,
-0.0215911865234375,
0.0111083984375,
-0.01093292236328125,
-0.031524658203125,
-0.0032978057861328125,
0.0243377685546875,
-0.0243988037109375,
-0.0609130859375,
0.011566162109375,
0.0232086181640625,
0.0197296142578125,
0.0328369140625,
-0.056793212890625,
0.0034542083740234375,
0.0019702911376953125,
-0.00988006591796875,
0.0274658203125,
0.046630859375,
-0.0013065338134765625,
0.043426513671875,
0.032623291015625,
0.025787353515625,
-0.0114593505859375,
0.005855560302734375,
0.0809326171875,
-0.0218658447265625,
-0.0257568359375,
-0.06597900390625,
0.038726806640625,
-0.0104827880859375,
-0.0286865234375,
0.042694091796875,
0.045257568359375,
0.03729248046875,
-0.0180511474609375,
0.032470703125,
0.006603240966796875,
0.0404052734375,
-0.01348114013671875,
0.062103271484375,
-0.0546875,
0.019683837890625,
-0.0148773193359375,
-0.051177978515625,
-0.01508331298828125,
0.046844482421875,
-0.022186279296875,
0.005336761474609375,
0.07720947265625,
0.056884765625,
-0.01039886474609375,
0.004505157470703125,
0.01221466064453125,
0.0283966064453125,
0.01885986328125,
0.04248046875,
0.00936126708984375,
-0.042144775390625,
0.0241851806640625,
-0.036163330078125,
-0.036590576171875,
-0.037200927734375,
-0.033416748046875,
-0.09234619140625,
-0.02252197265625,
-0.037506103515625,
-0.05120849609375,
-0.00963592529296875,
0.0704345703125,
0.0361328125,
-0.07464599609375,
0.0034542083740234375,
0.0162200927734375,
-0.0271453857421875,
-0.0188751220703125,
-0.014984130859375,
0.041595458984375,
0.01303863525390625,
-0.060760498046875,
-0.0014743804931640625,
0.0170135498046875,
0.005092620849609375,
-0.037445068359375,
0.0255889892578125,
-0.02508544921875,
-0.005401611328125,
0.0214385986328125,
0.0111083984375,
-0.04254150390625,
-0.01456451416015625,
-0.0207061767578125,
-0.046844482421875,
0.0012798309326171875,
0.05133056640625,
-0.06329345703125,
0.042205810546875,
0.01444244384765625,
-0.004016876220703125,
0.06866455078125,
-0.0163116455078125,
0.005222320556640625,
-0.0528564453125,
0.04736328125,
0.033935546875,
0.0247344970703125,
0.00444793701171875,
-0.02459716796875,
0.032684326171875,
0.036407470703125,
-0.0433349609375,
-0.06964111328125,
-0.01142120361328125,
-0.0947265625,
-0.0137176513671875,
0.0762939453125,
-0.00246429443359375,
-0.03509521484375,
0.0049896240234375,
-0.0290985107421875,
0.0236358642578125,
-0.0180511474609375,
0.057525634765625,
0.0247650146484375,
-0.0181732177734375,
-0.00814056396484375,
-0.060546875,
0.035858154296875,
0.00010794401168823242,
-0.017059326171875,
-0.0298614501953125,
0.019073486328125,
0.037567138671875,
0.030426025390625,
0.030670166015625,
-0.03070068359375,
0.0034942626953125,
-0.0015707015991210938,
0.0028285980224609375,
-0.0154876708984375,
-0.0204925537109375,
-0.01351165771484375,
0.0267486572265625,
-0.0087127685546875,
-0.017059326171875
]
] |
migtissera/Synthia-7B-v1.2 | 2023-10-14T01:33:37.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"arxiv:2306.02707",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | migtissera | null | null | migtissera/Synthia-7B-v1.2 | 10 | 7,638 | transformers | 2023-09-01T00:36:53 | ---
license: llama2
pipeline_tag: text-generation
language:
- en
library_name: transformers
---
Change from Synthia-7B -> Synthia-7B-v1.2: Capable of generalized Tree of Thought + Chain of Thought reasoning.
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-7B-v1.2
SynthIA (Synthetic Intelligent Agent) is a LLama-2-7B 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-7B-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|54.35|
|*hellaswag*|acc_norm|79.29|
|*mmlu*|acc_norm|49.33|
|*truthfulqa_mc*|mc2|48.92|
|**Total Average**|-|**57.97**||
<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-7B-v1.2"
output_file_path = "./Synthia-7B-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-7B-v1.2,
author = {Migel Tissera},
title = {Synthia-70B-v1.2b: 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}
}
```
| 5,544 | [
[
-0.0208587646484375,
-0.0738525390625,
0.034210205078125,
0.0183258056640625,
-0.0243072509765625,
0.00986480712890625,
-0.0177154541015625,
-0.046600341796875,
0.00678253173828125,
0.0172882080078125,
-0.053985595703125,
-0.04705810546875,
-0.0272979736328125,
0.003559112548828125,
-0.015045166015625,
0.0849609375,
-0.01108551025390625,
-0.01412200927734375,
-0.0027027130126953125,
-0.0005927085876464844,
-0.0310821533203125,
-0.047149658203125,
-0.050018310546875,
-0.032470703125,
0.0162353515625,
0.005466461181640625,
0.033203125,
0.049835205078125,
0.0179443359375,
0.0302276611328125,
-0.016265869140625,
0.0151824951171875,
-0.0246124267578125,
0.003726959228515625,
-0.00984954833984375,
-0.039764404296875,
-0.059478759765625,
0.0036144256591796875,
0.034393310546875,
0.020477294921875,
-0.003993988037109375,
0.034088134765625,
-0.003406524658203125,
0.0248565673828125,
-0.0248870849609375,
0.02142333984375,
-0.05145263671875,
-0.00887298583984375,
-0.0110931396484375,
-0.01157379150390625,
-0.0220794677734375,
-0.0214080810546875,
0.009735107421875,
-0.049560546875,
0.0089263916015625,
-0.001346588134765625,
0.0792236328125,
0.01459503173828125,
-0.0202484130859375,
-0.0290679931640625,
-0.040008544921875,
0.0556640625,
-0.0751953125,
0.01178741455078125,
0.0139312744140625,
0.00777435302734375,
-0.0207061767578125,
-0.061370849609375,
-0.06494140625,
-0.017333984375,
-0.00469970703125,
0.0166015625,
-0.0101165771484375,
0.004364013671875,
0.032928466796875,
0.027862548828125,
-0.041748046875,
-0.00860595703125,
-0.033905029296875,
-0.02069091796875,
0.046966552734375,
0.0203857421875,
0.027557373046875,
-0.0225372314453125,
-0.0221710205078125,
-0.025604248046875,
-0.042938232421875,
0.0234832763671875,
0.039337158203125,
0.019866943359375,
-0.02777099609375,
0.043609619140625,
-0.0159759521484375,
0.04742431640625,
0.01537322998046875,
-0.0134735107421875,
0.038421630859375,
-0.035247802734375,
-0.0308685302734375,
-0.01409149169921875,
0.0738525390625,
0.0149993896484375,
0.0159149169921875,
-0.0008087158203125,
-0.0020656585693359375,
0.01445770263671875,
-0.00365447998046875,
-0.061920166015625,
-0.0279541015625,
0.031341552734375,
-0.018096923828125,
-0.0244598388671875,
-0.01549530029296875,
-0.05548095703125,
-0.0018243789672851562,
-0.01678466796875,
0.02777099609375,
-0.025787353515625,
-0.034515380859375,
0.004589080810546875,
0.002780914306640625,
0.018768310546875,
0.004138946533203125,
-0.072509765625,
0.02716064453125,
0.0288238525390625,
0.06298828125,
0.004337310791015625,
-0.0268096923828125,
-0.00171661376953125,
-0.0035552978515625,
-0.00958251953125,
0.046295166015625,
-0.0266571044921875,
-0.0209197998046875,
-0.0280303955078125,
0.00347900390625,
-0.01971435546875,
-0.035675048828125,
0.0302276611328125,
-0.0210418701171875,
0.03594970703125,
-0.020263671875,
-0.033355712890625,
-0.0299072265625,
0.022796630859375,
-0.030181884765625,
0.08538818359375,
0.006793975830078125,
-0.06439208984375,
0.0036258697509765625,
-0.042724609375,
-0.010009765625,
-0.020294189453125,
-0.018218994140625,
-0.03765869140625,
-0.017822265625,
0.018829345703125,
0.022003173828125,
-0.0214080810546875,
0.033111572265625,
-0.021820068359375,
-0.014007568359375,
0.0256805419921875,
-0.02587890625,
0.085693359375,
0.01454925537109375,
-0.046875,
0.0242156982421875,
-0.06109619140625,
0.01861572265625,
0.0209808349609375,
-0.0261077880859375,
-0.005771636962890625,
-0.01422882080078125,
-0.0159454345703125,
0.0233001708984375,
0.0286102294921875,
-0.039764404296875,
0.023223876953125,
-0.047943115234375,
0.04052734375,
0.06170654296875,
0.004726409912109375,
0.0185089111328125,
-0.025726318359375,
0.03466796875,
0.00998687744140625,
0.005939483642578125,
0.0025806427001953125,
-0.047332763671875,
-0.07598876953125,
-0.00881195068359375,
0.00971221923828125,
0.050140380859375,
-0.03961181640625,
0.044525146484375,
-0.00952911376953125,
-0.0465087890625,
-0.039031982421875,
0.005489349365234375,
0.041412353515625,
0.04791259765625,
0.03656005859375,
-0.01397705078125,
-0.0601806640625,
-0.05902099609375,
-0.005466461181640625,
-0.0261993408203125,
0.003200531005859375,
0.01543426513671875,
0.05584716796875,
-0.0185394287109375,
0.06768798828125,
-0.032470703125,
0.002361297607421875,
-0.02520751953125,
0.0116119384765625,
0.036102294921875,
0.05743408203125,
0.03350830078125,
-0.029876708984375,
-0.0242156982421875,
-0.00909423828125,
-0.06854248046875,
-0.006847381591796875,
-0.0178070068359375,
-0.033935546875,
0.01093292236328125,
0.01142120361328125,
-0.072509765625,
0.023345947265625,
0.03302001953125,
-0.04144287109375,
0.048675537109375,
-0.00716400146484375,
-0.0019855499267578125,
-0.1026611328125,
0.0178680419921875,
-0.0058135986328125,
-0.0047454833984375,
-0.045379638671875,
0.0033111572265625,
-0.01073455810546875,
0.01617431640625,
-0.028961181640625,
0.044158935546875,
-0.032745361328125,
0.00524139404296875,
-0.0047454833984375,
0.016448974609375,
-0.003116607666015625,
0.072998046875,
-0.007568359375,
0.049957275390625,
0.043731689453125,
-0.043548583984375,
0.0323486328125,
0.022003173828125,
-0.0173797607421875,
0.021240234375,
-0.060455322265625,
0.0328369140625,
0.00684356689453125,
0.0272979736328125,
-0.0709228515625,
-0.0135498046875,
0.0506591796875,
-0.041015625,
0.0234375,
0.0026950836181640625,
-0.033660888671875,
-0.038848876953125,
-0.0173492431640625,
0.0343017578125,
0.038238525390625,
-0.035552978515625,
0.0465087890625,
0.0207977294921875,
0.0032863616943359375,
-0.045501708984375,
-0.048126220703125,
-0.01409149169921875,
-0.0268707275390625,
-0.04949951171875,
0.0208282470703125,
-0.0263824462890625,
-0.01403045654296875,
-0.006626129150390625,
-0.01056671142578125,
0.002166748046875,
0.005832672119140625,
0.0225372314453125,
0.033294677734375,
-0.006481170654296875,
0.006206512451171875,
0.004291534423828125,
0.0008921623229980469,
0.02789306640625,
-0.0138397216796875,
0.05352783203125,
-0.03228759765625,
-0.01296234130859375,
-0.042755126953125,
0.003871917724609375,
0.0367431640625,
-0.01320648193359375,
0.053314208984375,
0.037139892578125,
-0.028961181640625,
-0.00675201416015625,
-0.035430908203125,
-0.016693115234375,
-0.03875732421875,
0.0313720703125,
-0.039031982421875,
-0.048492431640625,
0.06414794921875,
0.007965087890625,
0.01207733154296875,
0.0555419921875,
0.056732177734375,
0.004535675048828125,
0.07049560546875,
0.0247344970703125,
0.00887298583984375,
0.0274658203125,
-0.05828857421875,
0.007843017578125,
-0.08343505859375,
-0.04058837890625,
-0.025390625,
-0.01519012451171875,
-0.043609619140625,
-0.0230255126953125,
0.00555419921875,
0.01197052001953125,
-0.047698974609375,
0.0309600830078125,
-0.056671142578125,
0.02392578125,
0.03753662109375,
0.0204620361328125,
0.017669677734375,
-0.01092529296875,
-0.0175323486328125,
0.011566162109375,
-0.05059814453125,
-0.046661376953125,
0.09466552734375,
0.03131103515625,
0.043975830078125,
0.002574920654296875,
0.057464599609375,
-0.00567626953125,
0.030181884765625,
-0.04425048828125,
0.05938720703125,
0.025665283203125,
-0.06427001953125,
-0.0171661376953125,
-0.035736083984375,
-0.06707763671875,
0.02764892578125,
-0.006893157958984375,
-0.081298828125,
0.007007598876953125,
0.0128021240234375,
-0.0310516357421875,
0.0285491943359375,
-0.0538330078125,
0.07110595703125,
-0.017425537109375,
-0.0243072509765625,
-0.00336456298828125,
-0.049560546875,
0.04534912109375,
0.006504058837890625,
0.007694244384765625,
-0.01059722900390625,
0.0164031982421875,
0.07830810546875,
-0.035614013671875,
0.07489013671875,
-0.006031036376953125,
-0.01148223876953125,
0.047332763671875,
-0.0030803680419921875,
0.04644775390625,
0.0020351409912109375,
0.0011873245239257812,
0.0143280029296875,
-0.0029201507568359375,
-0.018310546875,
-0.0423583984375,
0.058197021484375,
-0.078857421875,
-0.059844970703125,
-0.042755126953125,
-0.038055419921875,
-0.00022459030151367188,
0.021087646484375,
0.031982421875,
0.0286102294921875,
0.001220703125,
-0.0019483566284179688,
0.03741455078125,
-0.0135040283203125,
0.035430908203125,
0.022979736328125,
-0.0192108154296875,
-0.041473388671875,
0.05517578125,
0.0139923095703125,
0.0216827392578125,
0.0083465576171875,
0.00984954833984375,
-0.037017822265625,
-0.035614013671875,
-0.0372314453125,
0.03289794921875,
-0.0594482421875,
-0.02069091796875,
-0.060516357421875,
-0.0243377685546875,
-0.0238494873046875,
0.0089263916015625,
-0.0279388427734375,
-0.034332275390625,
-0.038055419921875,
-0.0270538330078125,
0.0309600830078125,
0.042999267578125,
0.00614166259765625,
0.016693115234375,
-0.031829833984375,
0.0163726806640625,
0.0251007080078125,
0.00850677490234375,
0.017181396484375,
-0.05230712890625,
-0.0159149169921875,
0.0251922607421875,
-0.043182373046875,
-0.0751953125,
0.0305938720703125,
-0.0008440017700195312,
0.043243408203125,
0.0069427490234375,
0.0009388923645019531,
0.062286376953125,
-0.0118560791015625,
0.0609130859375,
0.0128631591796875,
-0.08551025390625,
0.046478271484375,
-0.024932861328125,
0.027740478515625,
0.02099609375,
0.0106353759765625,
-0.0194549560546875,
-0.044342041015625,
-0.061248779296875,
-0.06671142578125,
0.053314208984375,
0.03851318359375,
0.00627899169921875,
-0.0029468536376953125,
0.0230712890625,
-0.00514984130859375,
0.0108184814453125,
-0.08782958984375,
-0.021148681640625,
-0.033477783203125,
-0.0218048095703125,
0.0150909423828125,
-0.006282806396484375,
-0.01548004150390625,
-0.037872314453125,
0.060455322265625,
-0.0009737014770507812,
0.0355224609375,
0.027435302734375,
-0.0013904571533203125,
-0.0165863037109375,
0.0218048095703125,
0.041839599609375,
0.0511474609375,
-0.0200653076171875,
0.00975799560546875,
0.03607177734375,
-0.034271240234375,
0.0168609619140625,
0.0020809173583984375,
-0.009246826171875,
-0.013763427734375,
0.034881591796875,
0.067626953125,
-0.006988525390625,
-0.033447265625,
0.0118255615234375,
0.006378173828125,
-0.0172119140625,
-0.0247039794921875,
0.0162353515625,
0.0199737548828125,
0.033203125,
0.018829345703125,
0.00884246826171875,
-0.002193450927734375,
-0.047882080078125,
-0.00811004638671875,
0.0243988037109375,
0.00908660888671875,
-0.045684814453125,
0.0621337890625,
0.01108551025390625,
-0.02099609375,
0.044830322265625,
-0.0170440673828125,
-0.048675537109375,
0.056884765625,
0.055450439453125,
0.07470703125,
-0.0054931640625,
0.01064300537109375,
0.045135498046875,
0.01995849609375,
-0.0018777847290039062,
0.035308837890625,
-0.004322052001953125,
-0.04986572265625,
-0.02716064453125,
-0.04718017578125,
-0.01126861572265625,
0.0286102294921875,
-0.027587890625,
-0.000072479248046875,
-0.043426513671875,
-0.03009033203125,
-0.012664794921875,
0.0177764892578125,
-0.052764892578125,
0.0243072509765625,
0.0057830810546875,
0.0511474609375,
-0.05633544921875,
0.0687255859375,
0.0406494140625,
-0.04425048828125,
-0.0872802734375,
-0.0091705322265625,
-0.01036834716796875,
-0.04913330078125,
0.047698974609375,
0.017425537109375,
-0.0191497802734375,
0.01122283935546875,
-0.0511474609375,
-0.0731201171875,
0.09356689453125,
0.02996826171875,
-0.032073974609375,
-0.017608642578125,
0.0037593841552734375,
0.056884765625,
-0.025970458984375,
0.040863037109375,
0.03948974609375,
0.031585693359375,
0.00897216796875,
-0.0648193359375,
0.038330078125,
-0.0362548828125,
-0.007579803466796875,
-0.0064239501953125,
-0.06488037109375,
0.08551025390625,
-0.0284576416015625,
-0.0305023193359375,
0.0179443359375,
0.057861328125,
0.041290283203125,
0.01995849609375,
0.02294921875,
0.0540771484375,
0.0634765625,
-0.0098876953125,
0.0657958984375,
-0.0263214111328125,
0.03857421875,
0.0819091796875,
0.006504058837890625,
0.0543212890625,
0.027099609375,
-0.025299072265625,
0.06646728515625,
0.06256103515625,
0.00036835670471191406,
0.0279388427734375,
0.0255126953125,
-0.00838470458984375,
-0.0128173828125,
0.010711669921875,
-0.04052734375,
0.0207366943359375,
0.022674560546875,
-0.0241241455078125,
0.004856109619140625,
-0.020111083984375,
0.017425537109375,
-0.01549530029296875,
0.003948211669921875,
0.043731689453125,
0.01161956787109375,
-0.0595703125,
0.081298828125,
-0.0111083984375,
0.04412841796875,
-0.04156494140625,
-0.0002834796905517578,
-0.0142669677734375,
0.01529693603515625,
-0.0227508544921875,
-0.048828125,
0.00811004638671875,
0.005603790283203125,
-0.0035076141357421875,
0.0013952255249023438,
0.032562255859375,
-0.035430908203125,
-0.034454345703125,
0.0189056396484375,
0.033447265625,
0.016876220703125,
0.01256561279296875,
-0.066162109375,
0.00927734375,
0.00937652587890625,
-0.0465087890625,
0.00928497314453125,
0.029022216796875,
0.007843017578125,
0.058197021484375,
0.05975341796875,
-0.00214385986328125,
0.00847625732421875,
-0.0146026611328125,
0.08404541015625,
-0.054168701171875,
-0.032562255859375,
-0.0782470703125,
0.04217529296875,
-0.007205963134765625,
-0.041473388671875,
0.06939697265625,
0.040283203125,
0.07037353515625,
-0.001590728759765625,
0.0565185546875,
-0.0169525146484375,
0.0165863037109375,
-0.04473876953125,
0.0435791015625,
-0.0287017822265625,
0.03167724609375,
-0.0217132568359375,
-0.07672119140625,
0.002025604248046875,
0.0599365234375,
-0.0267791748046875,
0.01511383056640625,
0.060546875,
0.061248779296875,
0.0002503395080566406,
-0.005397796630859375,
-0.0082855224609375,
0.034423828125,
0.034698486328125,
0.06646728515625,
0.054168701171875,
-0.039642333984375,
0.042083740234375,
-0.0279388427734375,
-0.0138702392578125,
0.0031261444091796875,
-0.0465087890625,
-0.08642578125,
-0.03973388671875,
-0.0226593017578125,
-0.04119873046875,
0.0008492469787597656,
0.08038330078125,
0.04986572265625,
-0.0625,
-0.0307769775390625,
-0.027984619140625,
0.01108551025390625,
-0.01548004150390625,
-0.0191802978515625,
0.0399169921875,
-0.00994110107421875,
-0.055908203125,
0.0169677734375,
-0.006175994873046875,
0.0300445556640625,
-0.0246429443359375,
-0.01654052734375,
-0.027740478515625,
0.00983428955078125,
0.023712158203125,
0.0289154052734375,
-0.064453125,
-0.0200958251953125,
0.0042724609375,
-0.014984130859375,
0.00998687744140625,
0.031768798828125,
-0.062286376953125,
0.0413818359375,
0.03961181640625,
0.011810302734375,
0.050994873046875,
-0.00018167495727539062,
0.03717041015625,
-0.037078857421875,
0.018035888671875,
0.0113983154296875,
0.023834228515625,
0.0212249755859375,
-0.029693603515625,
0.036865234375,
0.0308990478515625,
-0.041656494140625,
-0.05706787109375,
0.01319122314453125,
-0.07867431640625,
-0.01248931884765625,
0.0882568359375,
-0.023468017578125,
-0.02801513671875,
0.0036258697509765625,
-0.032806396484375,
0.052398681640625,
-0.0338134765625,
0.069580078125,
0.04443359375,
-0.01450347900390625,
-0.005397796630859375,
-0.023529052734375,
0.041961669921875,
0.0206451416015625,
-0.07330322265625,
-0.00211334228515625,
0.0224609375,
0.02532958984375,
0.02203369140625,
0.056854248046875,
0.01027679443359375,
0.00457763671875,
0.005096435546875,
0.002838134765625,
-0.0162811279296875,
-0.0168609619140625,
0.0006041526794433594,
-0.00623321533203125,
-0.01264190673828125,
-0.01611328125
]
] |
euclaise/falcon_1b_stage3 | 2023-09-20T01:11:17.000Z | [
"transformers",
"pytorch",
"falcon",
"text-generation",
"generated_from_trainer",
"custom_code",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | euclaise | null | null | euclaise/falcon_1b_stage3 | 0 | 7,636 | transformers | 2023-09-18T11:47:17 | ---
license: apache-2.0
base_model: euclaise/falcon_1b_stage2
tags:
- generated_from_trainer
model-index:
- name: falcon_1b_stage3
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. -->
# falcon_1b_stage3
This model is a fine-tuned version of [euclaise/falcon_1b_stage2](https://huggingface.co/euclaise/falcon_1b_stage2) on an unknown 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: 6e-06
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 128.0
- total_train_batch_size: 128.0
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.15
- num_epochs: 5
### Training results
### Framework versions
- Transformers 4.33.2
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3
| 1,184 | [
[
-0.040496826171875,
-0.045928955078125,
0.0009212493896484375,
0.0205230712890625,
-0.017425537109375,
-0.019561767578125,
0.00550079345703125,
-0.026397705078125,
0.024749755859375,
0.0311126708984375,
-0.06292724609375,
-0.033172607421875,
-0.05169677734375,
0.0021514892578125,
-0.0250091552734375,
0.0916748046875,
-0.0005192756652832031,
0.025787353515625,
0.003143310546875,
0.00798797607421875,
-0.015045166015625,
-0.036102294921875,
-0.08599853515625,
-0.045135498046875,
0.0312042236328125,
0.0260467529296875,
0.04766845703125,
0.07696533203125,
0.051605224609375,
0.0225067138671875,
-0.039398193359375,
-0.0022525787353515625,
-0.045318603515625,
-0.0272369384765625,
-0.004108428955078125,
-0.0302581787109375,
-0.058868408203125,
0.0014200210571289062,
0.05078125,
0.031829833984375,
-0.025054931640625,
0.0364990234375,
0.00485992431640625,
0.03729248046875,
-0.035064697265625,
0.04132080078125,
-0.038848876953125,
0.03485107421875,
-0.0163726806640625,
-0.0203704833984375,
-0.01454925537109375,
0.006740570068359375,
-0.01428985595703125,
-0.0625,
0.040985107421875,
-0.0061187744140625,
0.07843017578125,
0.0275421142578125,
-0.012359619140625,
-0.001972198486328125,
-0.0548095703125,
0.0283966064453125,
-0.041778564453125,
0.01352691650390625,
0.0235595703125,
0.052490234375,
0.0008687973022460938,
-0.07635498046875,
-0.0212860107421875,
-0.001251220703125,
0.005321502685546875,
0.01934814453125,
-0.00638580322265625,
0.0014400482177734375,
0.06024169921875,
0.01312255859375,
-0.032745361328125,
0.0126800537109375,
-0.05596923828125,
-0.0201416015625,
0.045989990234375,
0.0302276611328125,
-0.0139617919921875,
-0.0051727294921875,
-0.03857421875,
-0.0118255615234375,
-0.0222015380859375,
0.0297393798828125,
0.04205322265625,
0.035430908203125,
-0.015472412109375,
0.05023193359375,
-0.0239715576171875,
0.040557861328125,
0.0250244140625,
0.002773284912109375,
0.03814697265625,
0.00418853759765625,
-0.0298004150390625,
-0.0006794929504394531,
0.05035400390625,
0.0355224609375,
0.02630615234375,
0.0022735595703125,
-0.025115966796875,
-0.01525115966796875,
0.0251007080078125,
-0.07073974609375,
-0.007724761962890625,
0.00720977783203125,
-0.0323486328125,
-0.05108642578125,
0.0345458984375,
-0.04730224609375,
0.00315093994140625,
-0.011077880859375,
0.023162841796875,
-0.0070037841796875,
-0.0142364501953125,
0.0136260986328125,
-0.01070404052734375,
0.01282501220703125,
0.0022735595703125,
-0.0655517578125,
0.0399169921875,
0.03924560546875,
0.03704833984375,
0.01427459716796875,
-0.043975830078125,
-0.0386962890625,
0.01165771484375,
-0.024749755859375,
0.031951904296875,
-0.0197601318359375,
-0.041900634765625,
-0.005748748779296875,
0.0408935546875,
-0.015869140625,
-0.02996826171875,
0.0889892578125,
-0.024261474609375,
0.00728607177734375,
-0.03021240234375,
-0.04547119140625,
-0.0193328857421875,
0.0160064697265625,
-0.05230712890625,
0.0809326171875,
0.0177154541015625,
-0.0682373046875,
0.038543701171875,
-0.0650634765625,
-0.01476287841796875,
0.0115814208984375,
0.0021686553955078125,
-0.047698974609375,
-0.0027561187744140625,
0.007232666015625,
0.041229248046875,
-0.00370025634765625,
0.0159759521484375,
-0.037841796875,
-0.052276611328125,
-0.01088714599609375,
-0.018951416015625,
0.0479736328125,
0.03009033203125,
-0.024932861328125,
0.0018186569213867188,
-0.08447265625,
0.00797271728515625,
0.02484130859375,
-0.0261688232421875,
0.020111083984375,
-0.01971435546875,
0.037078857421875,
0.02362060546875,
0.02435302734375,
-0.04840087890625,
0.0079498291015625,
-0.025299072265625,
0.0270538330078125,
0.034393310546875,
0.00844573974609375,
0.00270843505859375,
-0.0367431640625,
0.0305023193359375,
0.031341552734375,
0.03271484375,
0.01251220703125,
-0.0310821533203125,
-0.07293701171875,
-0.0105133056640625,
0.01544952392578125,
0.02740478515625,
-0.018646240234375,
0.0438232421875,
0.0016384124755859375,
-0.050811767578125,
-0.01296234130859375,
-0.0009241104125976562,
0.0242767333984375,
0.04754638671875,
0.03314208984375,
-0.0008230209350585938,
-0.04022216796875,
-0.08197021484375,
-0.0055694580078125,
0.0021514892578125,
0.0252532958984375,
0.00827789306640625,
0.063232421875,
-0.0150909423828125,
0.040863037109375,
-0.030120849609375,
-0.007480621337890625,
-0.00800323486328125,
0.00209808349609375,
0.0252838134765625,
0.06597900390625,
0.0556640625,
-0.027435302734375,
-0.019073486328125,
-0.01102447509765625,
-0.052215576171875,
0.0232086181640625,
-0.00969696044921875,
-0.01262664794921875,
-0.00937652587890625,
0.03411865234375,
-0.037109375,
0.043365478515625,
0.0030612945556640625,
-0.022430419921875,
0.0230560302734375,
-0.031768798828125,
-0.004680633544921875,
-0.08172607421875,
0.01995849609375,
0.0271759033203125,
-0.00670623779296875,
-0.0250091552734375,
0.0208587646484375,
0.00017726421356201172,
-0.0113067626953125,
-0.0421142578125,
0.0462646484375,
-0.010498046875,
0.005035400390625,
-0.015380859375,
-0.03076171875,
-0.00014281272888183594,
0.050201416015625,
0.01617431640625,
0.043304443359375,
0.052978515625,
-0.03131103515625,
0.012725830078125,
0.03900146484375,
-0.01081085205078125,
0.0292205810546875,
-0.078857421875,
0.003536224365234375,
-0.0020580291748046875,
0.00746917724609375,
-0.053131103515625,
-0.028472900390625,
0.044708251953125,
-0.031524658203125,
0.0212249755859375,
-0.013885498046875,
-0.0275115966796875,
-0.0440673828125,
-0.00044536590576171875,
0.0179443359375,
0.029541015625,
-0.048065185546875,
0.021240234375,
-0.0023651123046875,
0.020843505859375,
-0.0360107421875,
-0.055694580078125,
-0.0238189697265625,
-0.01776123046875,
-0.035552978515625,
0.0169677734375,
-0.0012121200561523438,
0.01212310791015625,
-0.01529693603515625,
-0.0012340545654296875,
-0.018951416015625,
-0.0011425018310546875,
0.032928466796875,
0.00823211669921875,
-0.0106353759765625,
-0.01352691650390625,
0.01297760009765625,
-0.0244598388671875,
0.025665283203125,
-0.0038585662841796875,
0.01727294921875,
-0.01953125,
-0.018280029296875,
-0.062744140625,
0.00522613525390625,
0.049591064453125,
-0.01039886474609375,
0.053192138671875,
0.07037353515625,
-0.04388427734375,
-0.00792694091796875,
-0.0216064453125,
-0.01486968994140625,
-0.032989501953125,
0.039886474609375,
-0.04852294921875,
-0.02239990234375,
0.052490234375,
0.00608062744140625,
-0.001850128173828125,
0.0753173828125,
0.025360107421875,
0.004913330078125,
0.098388671875,
0.0189208984375,
0.00862884521484375,
0.0193328857421875,
-0.07147216796875,
-0.0271148681640625,
-0.06011962890625,
-0.042938232421875,
-0.0469970703125,
-0.0238037109375,
-0.050628662109375,
0.00374603271484375,
0.010284423828125,
0.019622802734375,
-0.0599365234375,
0.0258331298828125,
-0.03448486328125,
0.03900146484375,
0.040771484375,
0.0274810791015625,
-0.0081024169921875,
0.01320648193359375,
-0.02362060546875,
0.00028395652770996094,
-0.072509765625,
-0.02862548828125,
0.0880126953125,
0.04718017578125,
0.03363037109375,
-0.0242156982421875,
0.05279541015625,
-0.004589080810546875,
0.00002759695053100586,
-0.038970947265625,
0.0293426513671875,
-0.002285003662109375,
-0.06414794921875,
-0.0039520263671875,
-0.02349853515625,
-0.049530029296875,
0.00408935546875,
-0.035552978515625,
-0.04132080078125,
0.00537109375,
0.004497528076171875,
-0.0239410400390625,
0.02691650390625,
-0.032928466796875,
0.08172607421875,
-0.0123748779296875,
-0.0389404296875,
-0.00004315376281738281,
-0.0404052734375,
0.01506805419921875,
-0.0016908645629882812,
-0.004474639892578125,
0.01206207275390625,
0.00821685791015625,
0.0706787109375,
-0.058868408203125,
0.04827880859375,
-0.037139892578125,
0.033050537109375,
0.029144287109375,
-0.01082611083984375,
0.06207275390625,
0.0174713134765625,
-0.0209808349609375,
0.00757598876953125,
0.0165863037109375,
-0.051910400390625,
-0.030181884765625,
0.055572509765625,
-0.09014892578125,
-0.00742340087890625,
-0.04486083984375,
-0.033050537109375,
-0.0164031982421875,
0.01242828369140625,
0.044158935546875,
0.049835205078125,
-0.01511383056640625,
0.01580810546875,
0.0141143798828125,
0.0210113525390625,
0.0384521484375,
0.029266357421875,
-0.008453369140625,
-0.03802490234375,
0.0604248046875,
-0.0019178390502929688,
0.0004038810729980469,
-0.00801849365234375,
0.01184844970703125,
-0.033355712890625,
-0.045806884765625,
-0.0233306884765625,
0.0205841064453125,
-0.050384521484375,
0.00005632638931274414,
-0.0313720703125,
-0.042572021484375,
-0.00798797607421875,
-0.004405975341796875,
-0.036224365234375,
-0.033599853515625,
-0.055572509765625,
-0.01861572265625,
0.0213775634765625,
0.053466796875,
-0.00946044921875,
0.05999755859375,
-0.048126220703125,
-0.01560211181640625,
0.00543212890625,
0.031707763671875,
0.00014472007751464844,
-0.0672607421875,
-0.0299224853515625,
0.0231170654296875,
-0.0322265625,
-0.02618408203125,
0.02679443359375,
0.0090484619140625,
0.0572509765625,
0.0435791015625,
-0.022216796875,
0.06390380859375,
-0.013214111328125,
0.05804443359375,
0.0215606689453125,
-0.036712646484375,
0.03228759765625,
-0.035186767578125,
0.01467132568359375,
0.050994873046875,
0.041717529296875,
0.01007080078125,
-0.01412200927734375,
-0.09661865234375,
-0.03912353515625,
0.06292724609375,
0.032012939453125,
0.00414276123046875,
0.0014553070068359375,
0.053924560546875,
-0.0007038116455078125,
0.0117950439453125,
-0.0465087890625,
-0.0305023193359375,
-0.0296173095703125,
0.005950927734375,
-0.0225372314453125,
-0.01045989990234375,
-0.01161956787109375,
-0.057647705078125,
0.082275390625,
-0.00330352783203125,
0.01190948486328125,
0.0052642822265625,
0.01934814453125,
-0.0270538330078125,
-0.020538330078125,
0.054412841796875,
0.048828125,
-0.05023193359375,
-0.0248565673828125,
0.0074920654296875,
-0.03729248046875,
-0.0079803466796875,
0.022552490234375,
-0.0115203857421875,
0.0022869110107421875,
0.0254058837890625,
0.0758056640625,
0.01213836669921875,
0.0010042190551757812,
0.03009033203125,
-0.011016845703125,
-0.04241943359375,
-0.0191802978515625,
0.0289306640625,
-0.015625,
0.0290374755859375,
0.003143310546875,
0.0423583984375,
0.00909423828125,
-0.0047454833984375,
0.0216217041015625,
0.01512908935546875,
-0.048858642578125,
-0.01983642578125,
0.0738525390625,
0.01306915283203125,
-0.0321044921875,
0.040191650390625,
-0.01343536376953125,
-0.0128173828125,
0.07037353515625,
0.044952392578125,
0.0665283203125,
0.00539398193359375,
0.019927978515625,
0.07073974609375,
0.004726409912109375,
-0.02569580078125,
0.04241943359375,
0.00994110107421875,
-0.0347900390625,
0.0015382766723632812,
-0.050750732421875,
-0.007755279541015625,
0.031829833984375,
-0.0875244140625,
0.03558349609375,
-0.046600341796875,
-0.0312042236328125,
0.0198516845703125,
0.0265045166015625,
-0.06805419921875,
0.040802001953125,
0.0019369125366210938,
0.09814453125,
-0.06591796875,
0.055389404296875,
0.0513916015625,
-0.0469970703125,
-0.07891845703125,
-0.02166748046875,
-0.010894775390625,
-0.0703125,
0.0557861328125,
0.0006270408630371094,
0.01568603515625,
0.0264892578125,
-0.040740966796875,
-0.041015625,
0.077880859375,
0.011199951171875,
-0.05242919921875,
0.007617950439453125,
0.020843505859375,
0.04931640625,
-0.043243408203125,
0.054412841796875,
0.0181732177734375,
0.0278472900390625,
0.047393798828125,
-0.055267333984375,
-0.01788330078125,
-0.036773681640625,
0.0171356201171875,
-0.0013380050659179688,
-0.057830810546875,
0.07208251953125,
-0.00728607177734375,
0.0282440185546875,
0.022979736328125,
0.040771484375,
0.0124969482421875,
0.009002685546875,
0.0243988037109375,
0.055084228515625,
0.047760009765625,
-0.0011205673217773438,
0.06451416015625,
-0.059234619140625,
0.059417724609375,
0.07818603515625,
-0.0006794929504394531,
0.042266845703125,
0.033111572265625,
-0.00626373291015625,
0.011932373046875,
0.0777587890625,
-0.034393310546875,
0.028289794921875,
0.0173187255859375,
0.0118408203125,
-0.032470703125,
0.0174102783203125,
-0.059906005859375,
0.036712646484375,
-0.0010194778442382812,
-0.05364990234375,
-0.031463623046875,
-0.0236968994140625,
0.003391265869140625,
-0.016571044921875,
-0.038360595703125,
0.034149169921875,
-0.0301361083984375,
-0.045074462890625,
0.060760498046875,
0.00984954833984375,
0.014801025390625,
-0.044281005859375,
-0.0016584396362304688,
-0.017425537109375,
0.0261993408203125,
-0.0244903564453125,
-0.035858154296875,
0.0229034423828125,
-0.001483917236328125,
-0.0130462646484375,
0.005847930908203125,
0.021820068359375,
-0.016632080078125,
-0.08563232421875,
0.00566864013671875,
0.031890869140625,
0.020233154296875,
0.0038890838623046875,
-0.0870361328125,
0.004192352294921875,
-0.0157623291015625,
-0.0254364013671875,
0.0092620849609375,
0.0158233642578125,
0.018310546875,
0.037353515625,
0.0399169921875,
0.00348663330078125,
0.0002428293228149414,
0.01013946533203125,
0.0614013671875,
-0.033416748046875,
-0.0445556640625,
-0.044281005859375,
0.04144287109375,
-0.01007080078125,
-0.06854248046875,
0.0380859375,
0.0823974609375,
0.06634521484375,
-0.017974853515625,
0.04193115234375,
0.01560211181640625,
0.0308685302734375,
-0.031829833984375,
0.04608154296875,
-0.04638671875,
0.0068359375,
-0.022186279296875,
-0.064697265625,
0.004608154296875,
0.055084228515625,
-0.0028324127197265625,
0.0185546875,
0.0428466796875,
0.0723876953125,
-0.0204620361328125,
0.03472900390625,
0.0086517333984375,
0.00905609130859375,
0.01186370849609375,
0.03619384765625,
0.0406494140625,
-0.06817626953125,
0.020660400390625,
-0.04376220703125,
-0.007801055908203125,
-0.004528045654296875,
-0.0535888671875,
-0.081298828125,
-0.032623291015625,
-0.036285400390625,
-0.037841796875,
0.008636474609375,
0.072021484375,
0.06414794921875,
-0.055816650390625,
-0.03369140625,
-0.0220947265625,
-0.028533935546875,
-0.024749755859375,
-0.01222991943359375,
0.021087646484375,
-0.022003173828125,
-0.04949951171875,
-0.003925323486328125,
-0.024139404296875,
0.02569580078125,
-0.01461029052734375,
-0.023712158203125,
-0.01334381103515625,
-0.0171356201171875,
0.00945281982421875,
0.01490020751953125,
-0.038482666015625,
-0.02996826171875,
-0.00608062744140625,
-0.0029811859130859375,
0.012054443359375,
0.0219573974609375,
-0.0306854248046875,
0.0268402099609375,
0.00933074951171875,
0.01538848876953125,
0.074462890625,
0.0021915435791015625,
0.01543426513671875,
-0.0404052734375,
0.0289306640625,
0.0029277801513671875,
0.0313720703125,
-0.00850677490234375,
-0.028839111328125,
0.044342041015625,
0.035186767578125,
-0.033172607421875,
-0.050079345703125,
-0.0174560546875,
-0.0859375,
0.0099334716796875,
0.0848388671875,
0.003131866455078125,
-0.0335693359375,
0.03924560546875,
-0.02069091796875,
0.0255889892578125,
-0.017852783203125,
0.03363037109375,
0.0509033203125,
-0.00958251953125,
0.01280975341796875,
-0.042510986328125,
0.0396728515625,
-0.005504608154296875,
-0.054656982421875,
-0.0284881591796875,
0.0148468017578125,
0.039215087890625,
-0.01087188720703125,
0.013214111328125,
-0.0015878677368164062,
0.0236358642578125,
0.0171661376953125,
0.0033817291259765625,
-0.0562744140625,
-0.0350341796875,
-0.019134521484375,
0.01739501953125,
-0.00824737548828125,
-0.035369873046875
]
] |
dpv/finetuned-gpt2-tiny | 2023-06-23T15:12:26.000Z | [
"transformers",
"pytorch",
"tensorboard",
"gpt2",
"text-generation",
"generated_from_trainer",
"dataset:roneneldan/TinyStories",
"license:mit",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | dpv | null | null | dpv/finetuned-gpt2-tiny | 1 | 7,635 | transformers | 2023-06-23T14:58:18 | ---
license: mit
tags:
- generated_from_trainer
datasets:
- roneneldan/TinyStories
model-index:
- name: finetuned-gpt2-tiny
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. -->
# finetuned-gpt2-tiny
This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the roneneldan/TinyStories 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-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
### Framework versions
- Transformers 4.31.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
| 1,034 | [
[
-0.04010009765625,
-0.052459716796875,
0.018768310546875,
0.0006151199340820312,
-0.037017822265625,
-0.040313720703125,
-0.01500701904296875,
-0.019775390625,
0.00019288063049316406,
0.0183258056640625,
-0.0517578125,
-0.0195465087890625,
-0.04510498046875,
-0.009429931640625,
-0.00614166259765625,
0.107177734375,
0.005580902099609375,
0.03619384765625,
0.0004208087921142578,
0.0009617805480957031,
-0.029998779296875,
-0.0226593017578125,
-0.08392333984375,
-0.05841064453125,
0.029876708984375,
0.0267181396484375,
0.06561279296875,
0.07220458984375,
0.037017822265625,
0.017059326171875,
-0.02801513671875,
-0.01052093505859375,
-0.044464111328125,
-0.040496826171875,
-0.0136260986328125,
-0.030517578125,
-0.0657958984375,
0.006206512451171875,
0.042724609375,
0.0208587646484375,
-0.0187835693359375,
0.04608154296875,
0.0308685302734375,
0.01441192626953125,
-0.0165557861328125,
0.037353515625,
-0.044219970703125,
0.0234222412109375,
-0.00782012939453125,
-0.0039520263671875,
-0.01482391357421875,
-0.01355743408203125,
0.005535125732421875,
-0.04632568359375,
0.059844970703125,
-0.0026721954345703125,
0.07977294921875,
0.0298309326171875,
-0.01373291015625,
0.00415802001953125,
-0.06109619140625,
0.034515380859375,
-0.03521728515625,
0.01169586181640625,
0.0311737060546875,
0.045257568359375,
0.0102081298828125,
-0.07000732421875,
-0.0243682861328125,
-0.0074920654296875,
-0.0093536376953125,
0.007694244384765625,
-0.00836944580078125,
0.01052093505859375,
0.057647705078125,
0.028167724609375,
-0.052154541015625,
0.0119476318359375,
-0.041412353515625,
-0.01934814453125,
0.04241943359375,
0.03045654296875,
-0.0093536376953125,
-0.0173187255859375,
-0.027862548828125,
-0.01198577880859375,
-0.03924560546875,
-0.01461029052734375,
0.032012939453125,
0.025115966796875,
-0.034393310546875,
0.05291748046875,
-0.0095367431640625,
0.04052734375,
0.01157379150390625,
0.0038776397705078125,
0.01471710205078125,
-0.00485992431640625,
-0.0333251953125,
-0.00649261474609375,
0.0628662109375,
0.044158935546875,
0.0261993408203125,
-0.00724029541015625,
-0.030120849609375,
-0.00856781005859375,
0.0125274658203125,
-0.07867431640625,
-0.06134033203125,
-0.0008678436279296875,
-0.03741455078125,
-0.04071044921875,
0.003231048583984375,
-0.049713134765625,
0.00128936767578125,
-0.03826904296875,
0.051544189453125,
-0.017791748046875,
-0.0098876953125,
-0.007129669189453125,
-0.010650634765625,
0.0124359130859375,
0.0214996337890625,
-0.07330322265625,
0.037811279296875,
0.04559326171875,
0.048828125,
0.022216796875,
-0.031402587890625,
-0.013702392578125,
0.01007080078125,
-0.017242431640625,
0.0389404296875,
0.01001739501953125,
-0.035186767578125,
-0.022216796875,
0.02508544921875,
0.0015096664428710938,
-0.043060302734375,
0.064453125,
-0.0243377685546875,
0.022003173828125,
-0.0179595947265625,
-0.04107666015625,
-0.002056121826171875,
0.0203704833984375,
-0.04949951171875,
0.0792236328125,
0.0187530517578125,
-0.048583984375,
0.04583740234375,
-0.04547119140625,
-0.0082244873046875,
0.01148223876953125,
-0.00807952880859375,
-0.0628662109375,
0.004421234130859375,
-0.0024509429931640625,
0.02783203125,
-0.022705078125,
0.0234832763671875,
-0.0284576416015625,
-0.04046630859375,
-0.01213836669921875,
-0.05010986328125,
0.0400390625,
0.0190887451171875,
-0.02093505859375,
0.00997161865234375,
-0.0711669921875,
0.0259552001953125,
0.0266571044921875,
-0.03033447265625,
0.0106048583984375,
-0.0280303955078125,
0.043304443359375,
0.01947021484375,
0.01983642578125,
-0.034088134765625,
0.025115966796875,
-0.0288848876953125,
0.037506103515625,
0.051971435546875,
0.00832366943359375,
0.0118560791015625,
-0.0288848876953125,
0.00605010986328125,
-0.006214141845703125,
0.039337158203125,
0.05169677734375,
-0.042205810546875,
-0.06109619140625,
-0.01305389404296875,
0.03338623046875,
0.02703857421875,
-0.02783203125,
0.0421142578125,
-0.0118560791015625,
-0.05242919921875,
-0.019439697265625,
0.006591796875,
0.0301971435546875,
0.023040771484375,
0.037078857421875,
-0.01120758056640625,
-0.0282745361328125,
-0.071044921875,
-0.0076141357421875,
0.0017309188842773438,
-0.0111846923828125,
0.0182342529296875,
0.056488037109375,
-0.0192413330078125,
0.058746337890625,
-0.04638671875,
-0.0170135498046875,
-0.0129241943359375,
0.003757476806640625,
0.02093505859375,
0.06939697265625,
0.051116943359375,
-0.02313232421875,
-0.0299530029296875,
-0.0165252685546875,
-0.04400634765625,
0.019927978515625,
-0.0056304931640625,
-0.012603759765625,
-0.024261474609375,
0.015167236328125,
-0.06243896484375,
0.03369140625,
0.00917816162109375,
-0.0244598388671875,
0.034820556640625,
-0.0386962890625,
-0.0202789306640625,
-0.0804443359375,
-0.00150299072265625,
0.0175018310546875,
-0.01532745361328125,
-0.01244354248046875,
0.005214691162109375,
0.005695343017578125,
-0.02252197265625,
-0.02691650390625,
0.0511474609375,
-0.014404296875,
0.004482269287109375,
-0.0196075439453125,
-0.0171356201171875,
0.0014791488647460938,
0.0372314453125,
0.00704193115234375,
0.05572509765625,
0.040924072265625,
-0.03741455078125,
0.034820556640625,
0.04296875,
-0.0151519775390625,
0.0302276611328125,
-0.0784912109375,
0.018218994140625,
0.0037975311279296875,
-0.00732421875,
-0.04449462890625,
-0.03472900390625,
0.047607421875,
-0.024871826171875,
0.0382080078125,
-0.045928955078125,
-0.0457763671875,
-0.03668212890625,
-0.0003223419189453125,
0.024200439453125,
0.042083740234375,
-0.054443359375,
0.016143798828125,
-0.00257110595703125,
0.01102447509765625,
0.0013036727905273438,
-0.05859375,
-0.0221099853515625,
-0.0152435302734375,
-0.026641845703125,
0.01131439208984375,
0.0037384033203125,
0.00910186767578125,
-0.003173828125,
-0.005924224853515625,
-0.0225067138671875,
-0.01617431640625,
0.0214996337890625,
0.023162841796875,
-0.0134735107421875,
-0.0030536651611328125,
0.01031494140625,
-0.03369140625,
0.0218353271484375,
-0.01568603515625,
0.0301055908203125,
-0.008514404296875,
-0.00682830810546875,
-0.058349609375,
-0.0028247833251953125,
0.0223388671875,
0.007610321044921875,
0.05352783203125,
0.07745361328125,
-0.04180908203125,
-0.006134033203125,
-0.0186309814453125,
-0.03643798828125,
-0.03448486328125,
0.039794921875,
-0.036468505859375,
-0.03265380859375,
0.03204345703125,
0.006317138671875,
-0.01348876953125,
0.0701904296875,
0.0411376953125,
0.005046844482421875,
0.091064453125,
0.0235748291015625,
0.007785797119140625,
0.0296478271484375,
-0.061767578125,
-0.0102081298828125,
-0.061126708984375,
-0.019683837890625,
-0.029327392578125,
-0.0177764892578125,
-0.06427001953125,
-0.007183074951171875,
0.0173187255859375,
0.029693603515625,
-0.061859130859375,
0.036773681640625,
-0.05010986328125,
0.043304443359375,
0.05181884765625,
0.035797119140625,
-0.005615234375,
0.01120758056640625,
-0.00533294677734375,
-0.004497528076171875,
-0.06781005859375,
-0.03900146484375,
0.0965576171875,
0.0419921875,
0.05859375,
-0.001750946044921875,
0.041748046875,
-0.0008821487426757812,
0.00939178466796875,
-0.04449462890625,
0.031982421875,
-0.000476837158203125,
-0.06842041015625,
-0.01507568359375,
-0.02618408203125,
-0.044586181640625,
0.0072784423828125,
-0.0254364013671875,
-0.049285888671875,
0.0076141357421875,
0.03271484375,
-0.039093017578125,
0.018280029296875,
-0.06622314453125,
0.10089111328125,
-0.012420654296875,
-0.044158935546875,
-0.012359619140625,
-0.0350341796875,
0.005458831787109375,
0.0124053955078125,
-0.03167724609375,
0.0034580230712890625,
0.0189666748046875,
0.0665283203125,
-0.06622314453125,
0.052734375,
-0.03375244140625,
0.032623291015625,
0.035980224609375,
-0.006923675537109375,
0.061126708984375,
0.039398193359375,
0.0091705322265625,
0.0120086669921875,
0.009429931640625,
-0.04010009765625,
-0.0206756591796875,
0.06256103515625,
-0.092529296875,
-0.0038051605224609375,
-0.036407470703125,
-0.0241546630859375,
-0.01451873779296875,
0.01322174072265625,
0.0574951171875,
0.044891357421875,
-0.0114593505859375,
0.00033974647521972656,
0.03167724609375,
0.0257720947265625,
0.0252532958984375,
0.01461029052734375,
0.00939178466796875,
-0.0411376953125,
0.0740966796875,
-0.0014495849609375,
0.00824737548828125,
-0.01409912109375,
0.0240478515625,
-0.0307464599609375,
-0.0435791015625,
-0.03753662109375,
0.02545166015625,
-0.04150390625,
-0.0172576904296875,
-0.017608642578125,
-0.029327392578125,
-0.020263671875,
0.015777587890625,
-0.04412841796875,
-0.023651123046875,
-0.052337646484375,
-0.01404571533203125,
0.032989501953125,
0.05120849609375,
-0.017364501953125,
0.061492919921875,
-0.048919677734375,
0.0118255615234375,
0.03887939453125,
0.04437255859375,
-0.006534576416015625,
-0.05584716796875,
-0.0311279296875,
0.006786346435546875,
-0.03564453125,
-0.030517578125,
0.01491546630859375,
0.00843048095703125,
0.03448486328125,
0.031494140625,
-0.0229339599609375,
0.0521240234375,
-0.032684326171875,
0.053436279296875,
0.015106201171875,
-0.03204345703125,
0.036041259765625,
-0.0286102294921875,
0.00861358642578125,
0.048126220703125,
0.022705078125,
0.01348876953125,
0.005764007568359375,
-0.0882568359375,
-0.0482177734375,
0.06414794921875,
0.032806396484375,
0.00994110107421875,
0.01288604736328125,
0.046539306640625,
0.00325775146484375,
0.016143798828125,
-0.060516357421875,
-0.0267486572265625,
-0.021759033203125,
-0.006649017333984375,
-0.0183868408203125,
-0.035430908203125,
-0.0246124267578125,
-0.057037353515625,
0.07171630859375,
-0.0091705322265625,
0.041259765625,
-0.0004246234893798828,
0.010894775390625,
-0.015838623046875,
-0.00884246826171875,
0.03668212890625,
0.03564453125,
-0.046478271484375,
-0.0184173583984375,
0.0141754150390625,
-0.0509033203125,
-0.00787353515625,
0.0306396484375,
-0.022216796875,
0.00853729248046875,
0.01154327392578125,
0.08624267578125,
0.00479888916015625,
0.01561737060546875,
0.026763916015625,
-0.010498046875,
-0.0260162353515625,
-0.021636962890625,
0.02435302734375,
-0.02056884765625,
0.028717041015625,
-0.012481689453125,
0.0445556640625,
0.0104827880859375,
-0.007099151611328125,
0.017974853515625,
0.01404571533203125,
-0.043914794921875,
-0.0269317626953125,
0.0799560546875,
0.00861358642578125,
-0.017547607421875,
0.04986572265625,
-0.02545166015625,
-0.006999969482421875,
0.045379638671875,
0.0450439453125,
0.07012939453125,
0.00772857666015625,
0.0006465911865234375,
0.05450439453125,
0.0197906494140625,
-0.024871826171875,
0.030426025390625,
0.00018680095672607422,
-0.041748046875,
-0.01220703125,
-0.043670654296875,
-0.01727294921875,
0.034515380859375,
-0.0772705078125,
0.0362548828125,
-0.051910400390625,
-0.0384521484375,
0.0135650634765625,
0.0230255126953125,
-0.07745361328125,
0.041015625,
0.00543975830078125,
0.07861328125,
-0.06964111328125,
0.07952880859375,
0.0428466796875,
-0.038543701171875,
-0.08392333984375,
-0.004802703857421875,
0.00025272369384765625,
-0.05816650390625,
0.0299072265625,
-0.0109710693359375,
0.03179931640625,
0.033050537109375,
-0.03997802734375,
-0.03802490234375,
0.07867431640625,
0.0209503173828125,
-0.047882080078125,
-0.00681304931640625,
0.013397216796875,
0.045166015625,
-0.007137298583984375,
0.04290771484375,
0.031768798828125,
0.01861572265625,
0.008392333984375,
-0.06817626953125,
-0.0176849365234375,
-0.0096282958984375,
0.021148681640625,
0.0070953369140625,
-0.056243896484375,
0.076416015625,
-0.0030765533447265625,
0.02587890625,
0.03533935546875,
0.03179931640625,
0.005756378173828125,
0.00923919677734375,
0.01354217529296875,
0.05908203125,
0.039764404296875,
-0.032684326171875,
0.0726318359375,
-0.032379150390625,
0.08428955078125,
0.09930419921875,
0.0135650634765625,
0.03179931640625,
0.0262451171875,
-0.023468017578125,
-0.003650665283203125,
0.0687255859375,
-0.040924072265625,
0.049041748046875,
0.00617218017578125,
-0.00330352783203125,
-0.02532958984375,
0.02606201171875,
-0.056549072265625,
0.0179595947265625,
0.00511932373046875,
-0.048583984375,
-0.0266571044921875,
-0.0113525390625,
-0.006763458251953125,
-0.0286407470703125,
-0.0419921875,
0.05108642578125,
-0.01313018798828125,
-0.01169586181640625,
0.042327880859375,
0.0135040283203125,
0.024261474609375,
-0.04193115234375,
-0.0012950897216796875,
-0.005558013916015625,
0.03704833984375,
-0.01531982421875,
-0.057281494140625,
0.0157012939453125,
-0.00580596923828125,
-0.02301025390625,
-0.0018739700317382812,
0.044158935546875,
-0.02703857421875,
-0.049591064453125,
-0.0002372264862060547,
0.031646728515625,
0.01806640625,
-0.0107574462890625,
-0.0804443359375,
-0.0110015869140625,
-0.0072784423828125,
-0.0248870849609375,
0.0229339599609375,
0.0281982421875,
0.007965087890625,
0.0248565673828125,
0.031494140625,
0.004253387451171875,
-0.005458831787109375,
0.01561737060546875,
0.06573486328125,
-0.034088134765625,
-0.040557861328125,
-0.0533447265625,
0.0312042236328125,
-0.0135650634765625,
-0.063232421875,
0.0439453125,
0.0816650390625,
0.0760498046875,
-0.031463623046875,
0.035675048828125,
0.00830841064453125,
0.03558349609375,
-0.037933349609375,
0.060028076171875,
-0.023468017578125,
-0.0017652511596679688,
-0.0182037353515625,
-0.07818603515625,
0.0010318756103515625,
0.060943603515625,
-0.0247039794921875,
0.018768310546875,
0.03314208984375,
0.0618896484375,
-0.029510498046875,
0.0178680419921875,
0.01102447509765625,
0.011688232421875,
0.003246307373046875,
0.027801513671875,
0.034881591796875,
-0.07537841796875,
0.031982421875,
-0.052154541015625,
-0.0019817352294921875,
-0.007030487060546875,
-0.036651611328125,
-0.06988525390625,
-0.00937652587890625,
-0.04473876953125,
-0.0384521484375,
0.006069183349609375,
0.07196044921875,
0.06103515625,
-0.062469482421875,
-0.0137176513671875,
-0.021820068359375,
-0.019622802734375,
-0.01039886474609375,
-0.0169219970703125,
0.033203125,
0.0010156631469726562,
-0.053192138671875,
-0.00577545166015625,
-0.0296173095703125,
0.041046142578125,
-0.00020301342010498047,
-0.0188140869140625,
0.01482391357421875,
-0.03521728515625,
0.0221405029296875,
0.006214141845703125,
-0.0322265625,
-0.050537109375,
-0.03399658203125,
-0.0155792236328125,
0.00618743896484375,
0.0265350341796875,
-0.03515625,
0.0140838623046875,
0.005077362060546875,
0.0011472702026367188,
0.0537109375,
-0.0036525726318359375,
0.0166473388671875,
-0.059844970703125,
0.041656494140625,
0.01160430908203125,
0.0263519287109375,
0.002651214599609375,
-0.0266265869140625,
0.0447998046875,
0.044097900390625,
-0.05145263671875,
-0.048553466796875,
-0.01360321044921875,
-0.07110595703125,
0.0207061767578125,
0.08831787109375,
0.014007568359375,
-0.0197601318359375,
0.0260772705078125,
-0.032867431640625,
0.03125,
-0.01454925537109375,
0.0308990478515625,
0.03314208984375,
-0.004062652587890625,
-0.00350189208984375,
-0.05487060546875,
0.0521240234375,
0.0024013519287109375,
-0.039337158203125,
-0.013519287109375,
0.0303497314453125,
0.047149658203125,
-0.01068878173828125,
0.0213165283203125,
0.00011485815048217773,
0.027099609375,
0.0226287841796875,
0.01107025146484375,
-0.0269012451171875,
-0.03009033203125,
-0.0243377685546875,
0.0003235340118408203,
0.0212860107421875,
-0.036712646484375
]
] |
formulae/Dorflan | 2023-10-04T10:31:29.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"merge",
"slerp",
"en",
"dataset:Open-Orca/OpenOrca",
"dataset:conceptofmind/cot_submix_original",
"dataset:conceptofmind/t0_submix_original",
"dataset:conceptofmind/niv2_submix_original",
"dataset:conceptofmind/flan2021_submix_original",
"dataset:ehartford/dolphin",
"license:mit",
"text-generation-inference",
"region:us"
] | text-generation | formulae | null | null | formulae/Dorflan | 0 | 7,632 | transformers | 2023-10-03T09:07:03 | ---
license: mit
datasets:
- Open-Orca/OpenOrca
- conceptofmind/cot_submix_original
- conceptofmind/t0_submix_original
- conceptofmind/niv2_submix_original
- conceptofmind/flan2021_submix_original
- ehartford/dolphin
language:
- en
tags:
- merge
- slerp
inference: false
metrics:
- accuracy
- bleu
---
<h1 style="text-align: center">Dorflan</h1>
<h2 style="text-align: center">An experimental model</h2>
<hr>
| Model | Average ⬆️ | ARC | HellaSwag | MMLU | TruthfulQA |
|:------------:|:------------:|:-------:|:---------:|:-------:|:----------:|
| formulae/Dorflan 📑 | 58.19 | 54.44 | 75.78 | 51.36 | 51.17 |
## Model Details
Dorflan is an experimental merged model created from the following three foundation models:
- stabilityai/StableBeluga-7B
- ehartford/dolphin-llama2-7b
- AIDC-ai-business/Marcoroni-7B
Dorflan was created by merging the weights and architectures of these three models using a custom merging technique. No further fine-tuning was performed after the merge.
Once the model obtains it's evaluation scores, then we'll know if it works or not.
## Intended Use
As an experimental model, Dorflan is intended for testing and research purposes only. It should not be used for production systems or to generate content for public use.
## Training Data
Dorflan inherits training data from its three foundation models:
- StableBeluga-7B: COT, Niv2, t0, & FLAN2021
- dolphin-llama2-7b: Dolphin
- Marcoroni-7B: OpenOrca
## Limitations
As an untested merged model, Dorflan has unknown capabilities and limitations. Potential issues include:
- Instability due to merged architectures
- Compounded bias and issues from all three foundation models
- Decreased performance on some tasks compared to the foundation models
Extensive testing is required to characterize Dorflan's capabilities and limitations.
## Ethical Considerations
- Dorflan may exhibit harmful biases inherited from its training data
- Output may be unreliable or manipulated due to instability
- Experimental nature increases potential for misuse
Use this model ethically and do not deploy it for sensitive applications.
## Contact Information
Please report issues or concerns with this model to the creator for further investigation. | 2,270 | [
[
-0.02978515625,
-0.044189453125,
0.006256103515625,
0.017303466796875,
-0.0158843994140625,
-0.00395965576171875,
0.0002359151840209961,
-0.0577392578125,
0.002086639404296875,
0.0261383056640625,
-0.02459716796875,
-0.002727508544921875,
-0.03582763671875,
-0.0123748779296875,
-0.021575927734375,
0.10015869140625,
-0.03338623046875,
0.0184326171875,
0.0271759033203125,
-0.0199432373046875,
-0.01837158203125,
-0.0026988983154296875,
-0.046844482421875,
-0.03167724609375,
0.0281982421875,
0.033599853515625,
0.05426025390625,
0.037811279296875,
0.076904296875,
0.01343536376953125,
-0.03802490234375,
0.00021541118621826172,
-0.058929443359375,
-0.00212860107421875,
-0.0018291473388671875,
-0.00975799560546875,
-0.0809326171875,
0.0185089111328125,
0.0521240234375,
0.04931640625,
-0.0262908935546875,
0.0183258056640625,
-0.004756927490234375,
0.050323486328125,
-0.039947509765625,
-0.0018682479858398438,
-0.004062652587890625,
0.0178375244140625,
-0.0181884765625,
-0.0047149658203125,
0.0017442703247070312,
-0.024017333984375,
0.01934814453125,
-0.01898193359375,
0.018341064453125,
0.0193328857421875,
0.06695556640625,
0.020355224609375,
-0.03631591796875,
-0.0175018310546875,
-0.055328369140625,
0.06121826171875,
-0.051422119140625,
0.029022216796875,
0.003154754638671875,
0.02557373046875,
-0.00933837890625,
-0.040374755859375,
-0.05865478515625,
-0.044281005859375,
-0.00914764404296875,
0.00980377197265625,
-0.035614013671875,
-0.006427764892578125,
0.004665374755859375,
0.06976318359375,
-0.0430908203125,
0.01354217529296875,
-0.045806884765625,
0.0178070068359375,
0.04718017578125,
0.0159454345703125,
0.0092010498046875,
-0.005062103271484375,
-0.054473876953125,
-0.04400634765625,
-0.03924560546875,
0.0196380615234375,
0.0253448486328125,
0.0128326416015625,
-0.03631591796875,
0.0572509765625,
-0.03515625,
0.0408935546875,
0.01447296142578125,
-0.005199432373046875,
0.03729248046875,
-0.015655517578125,
-0.0504150390625,
0.01035308837890625,
0.05926513671875,
0.039093017578125,
-0.01393890380859375,
-0.01776123046875,
-0.019622802734375,
0.001766204833984375,
0.0191192626953125,
-0.041961669921875,
-0.004184722900390625,
0.02069091796875,
-0.0309295654296875,
-0.070556640625,
0.033111572265625,
-0.057403564453125,
0.004291534423828125,
0.00618743896484375,
0.034332275390625,
-0.014617919921875,
-0.050537109375,
0.01467132568359375,
0.00832366943359375,
0.0300445556640625,
-0.00021398067474365234,
-0.030792236328125,
0.040985107421875,
0.00750732421875,
0.0543212890625,
-0.033355712890625,
-0.027984619140625,
-0.0157318115234375,
0.0159912109375,
-0.033355712890625,
0.02752685546875,
0.01392364501953125,
-0.045013427734375,
-0.03900146484375,
0.001445770263671875,
-0.0004887580871582031,
-0.0257720947265625,
0.043060302734375,
-0.037384033203125,
0.00994873046875,
-0.0270538330078125,
-0.0002015829086303711,
-0.0188446044921875,
0.0278167724609375,
-0.0638427734375,
0.08868408203125,
0.01509857177734375,
-0.0750732421875,
0.0183258056640625,
-0.07720947265625,
-0.00977325439453125,
-0.0272369384765625,
0.009033203125,
-0.04193115234375,
0.004009246826171875,
-0.00443267822265625,
0.03192138671875,
-0.0411376953125,
0.038848876953125,
-0.0318603515625,
-0.018798828125,
0.0347900390625,
-0.0237884521484375,
0.06634521484375,
0.026641845703125,
-0.054595947265625,
-0.00591278076171875,
-0.0577392578125,
-0.0130462646484375,
0.0146026611328125,
-0.0273284912109375,
0.00623321533203125,
-0.03369140625,
0.007724761962890625,
0.00720977783203125,
0.01384735107421875,
-0.0279083251953125,
-0.0079803466796875,
-0.01324462890625,
-0.00934600830078125,
0.042938232421875,
0.00455474853515625,
0.029052734375,
-0.08172607421875,
0.043792724609375,
0.0237884521484375,
0.00305938720703125,
0.042083740234375,
-0.0061187744140625,
-0.07318115234375,
-0.0202789306640625,
0.0128173828125,
0.0277099609375,
-0.0137939453125,
0.018768310546875,
-0.0243988037109375,
-0.061004638671875,
0.0083770751953125,
-0.0034008026123046875,
0.045806884765625,
0.06719970703125,
0.038482666015625,
-0.0193328857421875,
-0.046112060546875,
-0.055572509765625,
0.0214080810546875,
-0.037567138671875,
0.037811279296875,
0.010009765625,
0.06439208984375,
-0.0007205009460449219,
0.054473876953125,
-0.032745361328125,
-0.0148162841796875,
-0.0162353515625,
0.0045013427734375,
0.028350830078125,
0.060638427734375,
0.0771484375,
-0.0643310546875,
-0.028533935546875,
0.00536346435546875,
-0.066650390625,
0.0037555694580078125,
0.025299072265625,
-0.035675048828125,
0.0046234130859375,
-0.0057830810546875,
-0.06402587890625,
0.05096435546875,
0.043365478515625,
-0.031951904296875,
0.046600341796875,
-0.0012111663818359375,
0.02972412109375,
-0.08319091796875,
0.0352783203125,
0.00441741943359375,
0.01529693603515625,
-0.04669189453125,
0.011688232421875,
-0.0144500732421875,
0.0006804466247558594,
-0.0335693359375,
0.044677734375,
-0.02142333984375,
-0.0123443603515625,
0.0006656646728515625,
-0.01471710205078125,
0.01131439208984375,
0.0233001708984375,
0.00225830078125,
0.043060302734375,
0.0263671875,
-0.053436279296875,
0.029510498046875,
0.0209808349609375,
-0.0161285400390625,
0.039703369140625,
-0.057098388671875,
-0.00818634033203125,
0.005641937255859375,
0.0199432373046875,
-0.07861328125,
-0.036834716796875,
0.033050537109375,
-0.0233612060546875,
0.023529052734375,
0.02276611328125,
-0.0030460357666015625,
-0.0267181396484375,
-0.038421630859375,
0.025054931640625,
0.035858154296875,
-0.0355224609375,
0.031463623046875,
0.01259613037109375,
0.01052093505859375,
-0.07916259765625,
-0.07476806640625,
-0.029052734375,
-0.03424072265625,
-0.05902099609375,
0.05859375,
-0.0099945068359375,
-0.045989990234375,
0.0261383056640625,
-0.02825927734375,
-0.03387451171875,
0.008148193359375,
0.01708984375,
0.033935546875,
-0.0289459228515625,
-0.02655029296875,
0.0140380859375,
0.0158233642578125,
-0.0149688720703125,
0.01120758056640625,
0.015869140625,
-0.0183258056640625,
-0.02099609375,
-0.038177490234375,
0.000942230224609375,
0.07305908203125,
-0.01690673828125,
0.0738525390625,
0.02178955078125,
-0.024871826171875,
0.00440216064453125,
-0.051910400390625,
0.0134429931640625,
-0.0267791748046875,
0.0201416015625,
-0.01418304443359375,
-0.04461669921875,
0.06561279296875,
0.040679931640625,
0.01404571533203125,
0.053375244140625,
0.053253173828125,
-0.0194854736328125,
0.03533935546875,
0.0477294921875,
0.0166778564453125,
0.028411865234375,
-0.0670166015625,
-0.00493621826171875,
-0.06390380859375,
-0.044342041015625,
-0.034393310546875,
-0.0273895263671875,
-0.0258941650390625,
-0.025634765625,
0.00879669189453125,
0.028472900390625,
-0.034088134765625,
0.0251007080078125,
-0.01500701904296875,
0.01258087158203125,
0.0181884765625,
0.0193939208984375,
0.0274658203125,
-0.007770538330078125,
-0.017974853515625,
0.0166015625,
-0.031951904296875,
-0.031524658203125,
0.0809326171875,
0.036041259765625,
0.0582275390625,
0.054229736328125,
0.050201416015625,
0.01316070556640625,
0.038177490234375,
-0.02423095703125,
0.0164031982421875,
0.006893157958984375,
-0.08245849609375,
-0.01021575927734375,
-0.040069580078125,
-0.04693603515625,
0.018798828125,
-0.0063629150390625,
-0.03265380859375,
0.0270538330078125,
0.0221710205078125,
-0.033416748046875,
0.032958984375,
-0.024444580078125,
0.05712890625,
-0.03900146484375,
-0.0286712646484375,
-0.00873565673828125,
-0.029876708984375,
0.042388916015625,
-0.006214141845703125,
0.001720428466796875,
-0.01224517822265625,
0.00724029541015625,
0.06085205078125,
-0.0498046875,
0.05670166015625,
-0.0086822509765625,
0.010986328125,
0.03387451171875,
0.005458831787109375,
0.0200042724609375,
0.006809234619140625,
-0.024932861328125,
0.008544921875,
-0.01641845703125,
-0.02081298828125,
-0.00616455078125,
0.05816650390625,
-0.08819580078125,
-0.0150909423828125,
-0.051788330078125,
-0.03729248046875,
-0.0128631591796875,
0.01947021484375,
0.017364501953125,
0.0300445556640625,
-0.041259765625,
0.030181884765625,
0.04364013671875,
0.01146697998046875,
0.049407958984375,
0.061279296875,
-0.0195465087890625,
-0.035797119140625,
0.0193328857421875,
0.01424407958984375,
0.0017881393432617188,
0.0003769397735595703,
0.011871337890625,
-0.032623291015625,
-0.051422119140625,
-0.032928466796875,
0.0186920166015625,
-0.038848876953125,
-0.032196044921875,
-0.024993896484375,
-0.0170440673828125,
-0.0228424072265625,
-0.01067352294921875,
-0.0199432373046875,
-0.032073974609375,
-0.0296478271484375,
-0.042144775390625,
0.033599853515625,
0.06365966796875,
-0.02264404296875,
0.054290771484375,
-0.025634765625,
0.0180206298828125,
0.0245513916015625,
0.04156494140625,
0.01033782958984375,
-0.06414794921875,
-0.007549285888671875,
0.01371002197265625,
-0.036376953125,
-0.07672119140625,
0.0225982666015625,
-0.01320648193359375,
0.06671142578125,
0.01203155517578125,
0.00926971435546875,
0.0552978515625,
0.001728057861328125,
0.04815673828125,
0.02911376953125,
-0.062164306640625,
0.02972412109375,
-0.041168212890625,
-0.0184326171875,
0.040924072265625,
0.040924072265625,
0.00933074951171875,
-0.006755828857421875,
-0.05499267578125,
-0.039947509765625,
0.07916259765625,
0.0295257568359375,
-0.01507568359375,
0.0034637451171875,
0.02642822265625,
-0.0217742919921875,
0.01708984375,
-0.093994140625,
-0.016845703125,
-0.00809478759765625,
-0.01120758056640625,
0.0117034912109375,
-0.03515625,
-0.02642822265625,
-0.00968170166015625,
0.061279296875,
0.030914306640625,
0.0024662017822265625,
0.013885498046875,
-0.021697998046875,
-0.0367431640625,
-0.01221466064453125,
0.03753662109375,
0.07122802734375,
-0.044952392578125,
-0.00970458984375,
0.0276947021484375,
-0.0361328125,
0.0234832763671875,
0.01282501220703125,
0.0186767578125,
-0.021820068359375,
0.03228759765625,
0.068359375,
0.019287109375,
-0.02618408203125,
0.031707763671875,
0.002704620361328125,
-0.0292205810546875,
-0.0251007080078125,
0.0157623291015625,
-0.0193634033203125,
0.0281829833984375,
0.028564453125,
0.025543212890625,
-0.005306243896484375,
-0.019012451171875,
0.01282501220703125,
0.006229400634765625,
-0.031524658203125,
-0.012054443359375,
0.05072021484375,
0.00640106201171875,
-0.0036334991455078125,
0.029693603515625,
-0.005054473876953125,
-0.00753021240234375,
0.052520751953125,
0.040802001953125,
0.044891357421875,
-0.021240234375,
0.0106048583984375,
0.0101776123046875,
0.01251220703125,
-0.045928955078125,
0.036865234375,
0.017242431640625,
-0.054473876953125,
-0.0272064208984375,
-0.058837890625,
-0.03314208984375,
0.00888824462890625,
-0.07952880859375,
0.0235137939453125,
-0.042633056640625,
-0.0307769775390625,
-0.012786865234375,
0.0287017822265625,
-0.04180908203125,
0.0474853515625,
-0.00023114681243896484,
0.06689453125,
-0.08026123046875,
0.04949951171875,
0.070556640625,
-0.058380126953125,
-0.083984375,
-0.00623321533203125,
-0.00942230224609375,
-0.045440673828125,
0.0672607421875,
0.0032672882080078125,
-0.021728515625,
-0.0208587646484375,
-0.019866943359375,
-0.07867431640625,
0.11712646484375,
0.037933349609375,
-0.0204010009765625,
-0.0002548694610595703,
-0.013885498046875,
0.0404052734375,
-0.02789306640625,
0.021331787109375,
0.031005859375,
0.023162841796875,
0.0300140380859375,
-0.0828857421875,
0.003208160400390625,
-0.012420654296875,
0.003467559814453125,
-0.00164794921875,
-0.082275390625,
0.09307861328125,
-0.01116943359375,
0.02117919921875,
0.00278472900390625,
0.04241943359375,
0.045379638671875,
0.023345947265625,
0.0357666015625,
0.07452392578125,
0.051788330078125,
0.003421783447265625,
0.07318115234375,
-0.0212860107421875,
0.0516357421875,
0.0660400390625,
-0.0072174072265625,
0.0557861328125,
0.0269622802734375,
-0.02752685546875,
0.038482666015625,
0.041595458984375,
-0.000021219253540039062,
0.042694091796875,
-0.006725311279296875,
-0.0029621124267578125,
0.0248260498046875,
0.01079559326171875,
-0.056976318359375,
0.043609619140625,
-0.005008697509765625,
-0.03350830078125,
-0.00597381591796875,
-0.032928466796875,
0.014007568359375,
-0.029266357421875,
-0.01338958740234375,
0.008697509765625,
-0.004512786865234375,
-0.037933349609375,
0.032073974609375,
0.0092010498046875,
0.03302001953125,
-0.0390625,
0.00992584228515625,
-0.0299072265625,
0.01538848876953125,
-0.01291656494140625,
-0.0347900390625,
0.033447265625,
0.021484375,
-0.021240234375,
-0.005733489990234375,
0.01739501953125,
0.00028586387634277344,
-0.044677734375,
0.02978515625,
0.03778076171875,
0.0075531005859375,
0.04022216796875,
-0.0665283203125,
0.029266357421875,
0.0020160675048828125,
-0.049468994140625,
-0.0007281303405761719,
0.0096282958984375,
0.006855010986328125,
0.0357666015625,
0.046295166015625,
0.009521484375,
0.0084228515625,
-0.033447265625,
0.0894775390625,
-0.036285400390625,
-0.033172607421875,
-0.070068359375,
0.060791015625,
-0.0164947509765625,
-0.046722412109375,
0.050201416015625,
0.0345458984375,
0.054718017578125,
0.015899658203125,
0.03656005859375,
-0.040496826171875,
0.02587890625,
-0.04486083984375,
0.066650390625,
-0.05377197265625,
0.0190277099609375,
-0.030609130859375,
-0.08251953125,
-0.01374053955078125,
0.0655517578125,
-0.0257110595703125,
0.01515960693359375,
0.0430908203125,
0.05078125,
-0.0247039794921875,
0.0048828125,
0.0025539398193359375,
0.02447509765625,
0.0128936767578125,
0.022247314453125,
0.05828857421875,
-0.032745361328125,
0.043792724609375,
-0.048126220703125,
-0.0298309326171875,
-0.031463623046875,
-0.06842041015625,
-0.074951171875,
-0.0032196044921875,
-0.0228424072265625,
-0.0438232421875,
0.00597381591796875,
0.04339599609375,
0.052398681640625,
-0.067138671875,
-0.046722412109375,
-0.00815582275390625,
-0.007198333740234375,
-0.00848388671875,
-0.0142364501953125,
0.0251007080078125,
0.004169464111328125,
-0.0269775390625,
0.0282440185546875,
-0.004062652587890625,
0.052490234375,
-0.036102294921875,
-0.019012451171875,
-0.01514434814453125,
-0.021514892578125,
-0.01690673828125,
0.01812744140625,
-0.04107666015625,
-0.0033893585205078125,
0.0081939697265625,
-0.00833892822265625,
-0.00390625,
0.0247955322265625,
-0.0231475830078125,
0.00846099853515625,
0.036590576171875,
0.0171356201171875,
0.05474853515625,
-0.0030765533447265625,
0.04107666015625,
-0.0379638671875,
0.032196044921875,
0.01189422607421875,
0.036834716796875,
0.01324462890625,
-0.025146484375,
0.05474853515625,
0.00563812255859375,
-0.0283355712890625,
-0.0999755859375,
-0.0003478527069091797,
-0.07537841796875,
0.00333404541015625,
0.049957275390625,
-0.0188140869140625,
-0.01061248779296875,
0.0350341796875,
0.0088958740234375,
0.031341552734375,
-0.01325225830078125,
0.040557861328125,
0.040496826171875,
-0.040283203125,
0.00806427001953125,
-0.04388427734375,
0.018402099609375,
0.02496337890625,
-0.034576416015625,
-0.017578125,
0.03851318359375,
0.041259765625,
0.01448822021484375,
0.055816650390625,
-0.02410888671875,
0.0274658203125,
0.00890350341796875,
-0.007354736328125,
-0.02069091796875,
-0.0338134765625,
-0.02496337890625,
0.00616455078125,
0.018463134765625,
-0.01519775390625
]
] |
TheBloke/OpenOrca-Platypus2-13B-GPTQ | 2023-09-27T12:45:43.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"en",
"dataset:garage-bAInd/Open-Platypus",
"dataset:Open-Orca/OpenOrca",
"arxiv:2308.07317",
"arxiv:2306.02707",
"arxiv:2301.13688",
"license:cc-by-nc-4.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/OpenOrca-Platypus2-13B-GPTQ | 50 | 7,631 | transformers | 2023-08-12T19:45:04 | ---
language:
- en
license: cc-by-nc-4.0
library_name: transformers
datasets:
- garage-bAInd/Open-Platypus
- Open-Orca/OpenOrca
model_name: OpenOrca Platypus2 13B
base_model: Open-Orca/OpenOrca-Platypus2-13B
inference: false
model_creator: Open-Orca
model_type: llama
pipeline_tag: text-generation
prompt_template: '### Instruction:
{prompt}
### Response:
'
quantized_by: TheBloke
---
<!-- 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 -->
# OpenOrca Platypus2 13B - GPTQ
- Model creator: [Open-Orca](https://huggingface.co/Open-Orca)
- Original model: [OpenOrca Platypus2 13B](https://huggingface.co/Open-Orca/OpenOrca-Platypus2-13B)
<!-- description start -->
## Description
This repo contains GPTQ model files for [Open-Orca's OpenOrca Platypus2 13B](https://huggingface.co/Open-Orca/OpenOrca-Platypus2-13B).
Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
<!-- description end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GGUF)
* [Open-Orca's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/Open-Orca/OpenOrca-Platypus2-13B)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: Alpaca-InstructOnly
```
### Instruction:
{prompt}
### Response:
```
<!-- prompt-template end -->
<!-- licensing start -->
## Licensing
The creator of the source model has listed its license as `cc-by-nc-4.0`, and this quantization has therefore used that same license.
As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.
In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: [Open-Orca's OpenOrca Platypus2 13B](https://huggingface.co/Open-Orca/OpenOrca-Platypus2-13B).
<!-- licensing end -->
<!-- README_GPTQ.md-provided-files start -->
## Provided files and GPTQ parameters
Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.
Each separate quant is in a different branch. See below for instructions on fetching from different branches.
All recent GPTQ files are made with AutoGPTQ, and all files in non-main branches are made with AutoGPTQ. Files in the `main` branch which were uploaded before August 2023 were made with GPTQ-for-LLaMa.
<details>
<summary>Explanation of GPTQ parameters</summary>
- Bits: The bit size of the quantised model.
- GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
- Act Order: True or False. Also known as `desc_act`. True results in better quantisation accuracy. Some GPTQ clients have had issues with models that use Act Order plus Group Size, but this is generally resolved now.
- Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
- GPTQ dataset: The dataset used for quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
- Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
- ExLlama Compatibility: Whether this file can be loaded with ExLlama, which currently only supports Llama models in 4-bit.
</details>
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
| [main](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ/tree/main) | 4 | 128 | No | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 7.26 GB | Yes | 4-bit, without Act Order and group size 128g. |
| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 8.00 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
| [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 7.51 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
| [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 7.26 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
| [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 13.36 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. |
| [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 13.65 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
<!-- README_GPTQ.md-provided-files end -->
<!-- README_GPTQ.md-download-from-branches start -->
## How to download from branches
- In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/OpenOrca-Platypus2-13B-GPTQ:main`
- With Git, you can clone a branch with:
```
git clone --single-branch --branch main https://huggingface.co/TheBloke/OpenOrca-Platypus2-13B-GPTQ
```
- In Python Transformers code, the branch is the `revision` parameter; see below.
<!-- README_GPTQ.md-download-from-branches end -->
<!-- README_GPTQ.md-text-generation-webui start -->
## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.
1. Click the **Model tab**.
2. Under **Download custom model or LoRA**, enter `TheBloke/OpenOrca-Platypus2-13B-GPTQ`.
- To download from a specific branch, enter for example `TheBloke/OpenOrca-Platypus2-13B-GPTQ:main`
- see Provided Files above for the list of branches for each option.
3. Click **Download**.
4. The model will start downloading. Once it's finished it will say "Done".
5. In the top left, click the refresh icon next to **Model**.
6. In the **Model** dropdown, choose the model you just downloaded: `OpenOrca-Platypus2-13B-GPTQ`
7. The model will automatically load, and is now ready for use!
8. If you want any custom settings, set them and then click **Save settings for this model** followed by **Reload the Model** in the top right.
* Note that you do not need to and should not set manual GPTQ parameters any more. These are set automatically from the file `quantize_config.json`.
9. Once you're ready, click the **Text Generation tab** and enter a prompt to get started!
<!-- README_GPTQ.md-text-generation-webui end -->
<!-- README_GPTQ.md-use-from-python start -->
## How to use this GPTQ model from Python code
### Install the necessary packages
Requires: Transformers 4.32.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.
```shell
pip3 install transformers>=4.32.0 optimum>=1.12.0
pip3 install auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/ # Use cu117 if on CUDA 11.7
```
If you have problems installing AutoGPTQ using the pre-built wheels, install it from source instead:
```shell
pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
pip3 install .
```
### For CodeLlama models only: you must use Transformers 4.33.0 or later.
If 4.33.0 is not yet released when you read this, you will need to install Transformers from source:
```shell
pip3 uninstall -y transformers
pip3 install git+https://github.com/huggingface/transformers.git
```
### You can then use the following code
```python
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
model_name_or_path = "TheBloke/OpenOrca-Platypus2-13B-GPTQ"
# To use a different branch, change revision
# For example: revision="main"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
device_map="auto",
trust_remote_code=False,
revision="main")
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
prompt = "Tell me about AI"
prompt_template=f'''### Instruction:
{prompt}
### Response:
'''
print("\n\n*** Generate:")
input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
print(tokenizer.decode(output[0]))
# Inference can also be done using transformers' pipeline
print("*** Pipeline:")
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
max_new_tokens=512,
do_sample=True,
temperature=0.7,
top_p=0.95,
top_k=40,
repetition_penalty=1.1
)
print(pipe(prompt_template)[0]['generated_text'])
```
<!-- README_GPTQ.md-use-from-python end -->
<!-- README_GPTQ.md-compatibility start -->
## Compatibility
The files provided are tested to work with AutoGPTQ, both via Transformers and using AutoGPTQ directly. They should also work with [Occ4m's GPTQ-for-LLaMa fork](https://github.com/0cc4m/KoboldAI).
[ExLlama](https://github.com/turboderp/exllama) is compatible with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
[Huggingface Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) is compatible with all GPTQ models.
<!-- README_GPTQ.md-compatibility end -->
<!-- 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!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
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**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
# Original model card: Open-Orca's OpenOrca Platypus2 13B
<p><h1>🐋 The First OrcaPlatypus! 🐋</h1></p>
# OpenOrca-Platypus2-13B
OpenOrca-Platypus2-13B is a merge of [`garage-bAInd/Platypus2-13B`](https://huggingface.co/garage-bAInd/Platypus2-13B) and [`Open-Orca/OpenOrcaxOpenChat-Preview2-13B`](https://huggingface.co/Open-Orca/OpenOrcaxOpenChat-Preview2-13B).
This model is more than the sum of its parts! We are happy to be teaming up with the [Platypus](https://platypus-llm.github.io/) team to bring you a new model which once again tops the leaderboards!
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
## HuggingFace Leaderboard Performance
| Metric | Value |
|-----------------------|-------|
| MMLU (5-shot) | 59.5 |
| ARC (25-shot) | 62.88 |
| HellaSwag (10-shot) | 83.19 |
| TruthfulQA (0-shot) | 52.69 |
| Avg. | 64.56 |
We use [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.
## AGIEval Performance
We compare our results to our base Preview2 model (using LM Evaluation Harness).
We find **112%** of the base model's performance on AGI Eval, averaging **0.463**.
A large part of this boost is the substantial improvement to LSAT Logical Reasoning performance.
## BigBench-Hard Performance
We compare our results to our base Preview2 model (using LM Evaluation Harness).
We find **105%** of the base model's performance on BigBench-Hard, averaging **0.442**.
# Model Details
* **Trained by**: **Platypus2-13B** trained by Cole Hunter & Ariel Lee; **OpenOrcaxOpenChat-Preview2-13B** trained by Open-Orca
* **Model type:** **OpenOrca-Platypus2-13B** is an auto-regressive language model based on the Lllama 2 transformer architecture.
* **Language(s)**: English
* **License for Platypus2-13B base weights**: Non-Commercial Creative Commons license ([CC BY-NC-4.0](https://creativecommons.org/licenses/by-nc/4.0/))
* **License for OpenOrcaxOpenChat-Preview2-13B base weights**: Llama 2 Commercial
# Prompting
## Prompt Template for base Platypus2-13B
```
### Instruction:
<prompt> (without the <>)
### Response:
```
## Prompt Template for base OpenOrcaxOpenChat-Preview2-13B
OpenChat Llama2 V1: see [OpenOrcaxOpenChat-Preview2-13B](https://huggingface.co/Open-Orca/OpenOrcaxOpenChat-Preview2-13B) for additional information.
# Training
## Training Datasets
`garage-bAInd/Platypus2-13B` 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.
`Open-Orca/OpenOrcaxOpenChat-Preview2-13B` trained using a refined subset of most of the GPT-4 data from the [OpenOrca dataset](https://huggingface.co/datasets/Open-Orca/OpenOrca).
## Training Procedure
`Open-Orca/Platypus2-13B` was instruction fine-tuned using LoRA on 1x A100-80GB.
For training details and inference instructions please see the [Platypus](https://github.com/arielnlee/Platypus) GitHub repo.
# Supplemental
## Reproducing Evaluation Results (for HuggingFace Leaderboard Eval)
Install LM Evaluation Harness:
```
# clone repository
git clone https://github.com/EleutherAI/lm-evaluation-harness.git
# change to repo directory
cd lm-evaluation-harness
# check out the correct commit
git checkout b281b0921b636bc36ad05c0b0b0763bd6dd43463
# 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=Open-Orca/OpenOrca-Platypus2-13B --tasks arc_challenge --batch_size 1 --no_cache --write_out --output_path results/OpenOrca-Platypus2-13B/arc_challenge_25shot.json --device cuda --num_fewshot 25
```
HellaSwag:
```
python main.py --model hf-causal-experimental --model_args pretrained=Open-Orca/OpenOrca-Platypus2-13B --tasks hellaswag --batch_size 1 --no_cache --write_out --output_path results/OpenOrca-Platypus2-13B/hellaswag_10shot.json --device cuda --num_fewshot 10
```
MMLU:
```
python main.py --model hf-causal-experimental --model_args pretrained=Open-Orca/OpenOrca-Platypus2-13B --tasks hendrycksTest-* --batch_size 1 --no_cache --write_out --output_path results/OpenOrca-Platypus2-13B/mmlu_5shot.json --device cuda --num_fewshot 5
```
TruthfulQA:
```
python main.py --model hf-causal-experimental --model_args pretrained=Open-Orca/OpenOrca-Platypus2-13B --tasks truthfulqa_mc --batch_size 1 --no_cache --write_out --output_path results/OpenOrca-Platypus2-13B/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
@software{hunterlee2023orcaplaty1
title = {OpenOrcaPlatypus: Llama2-13B Model Instruct-tuned on Filtered OpenOrcaV1 GPT-4 Dataset and Merged with divergent STEM and Logic Dataset Model},
author = {Ariel N. Lee and Cole J. Hunter and Nataniel Ruiz and Bleys Goodson and Wing Lian and Guan Wang and Eugene Pentland and Austin Cook and Chanvichet Vong and "Teknium"},
year = {2023},
publisher = {HuggingFace},
journal = {HuggingFace repository},
howpublished = {\url{https://huggingface.co/Open-Orca/OpenOrca-Platypus2-13B},
}
@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}
}
@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{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= arXiv 2307.09288
}
@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}
}
@article{hu2021lora,
title={LoRA: Low-Rank Adaptation of Large Language Models},
author={Hu, Edward J. and Shen, Yelong and Wallis, Phillip and Allen-Zhu, Zeyuan and Li, Yuanzhi and Wang, Shean and Chen, Weizhu},
journal={CoRR},
year={2021}
}
```
| 26,299 | [
[
-0.0399169921875,
-0.054931640625,
0.007266998291015625,
0.0175933837890625,
-0.028106689453125,
-0.01253509521484375,
0.00047087669372558594,
-0.047943115234375,
0.02154541015625,
0.0300140380859375,
-0.036346435546875,
-0.040802001953125,
-0.024658203125,
-0.005687713623046875,
-0.0131683349609375,
0.083984375,
0.001758575439453125,
-0.0218505859375,
-0.0019102096557617188,
-0.03826904296875,
-0.0224456787109375,
-0.0347900390625,
-0.058624267578125,
-0.00856781005859375,
0.03826904296875,
0.01012420654296875,
0.0711669921875,
0.04559326171875,
0.01739501953125,
0.02008056640625,
-0.0164031982421875,
0.0016584396362304688,
-0.03729248046875,
-0.016998291015625,
0.00870513916015625,
-0.0254364013671875,
-0.04833984375,
0.0085296630859375,
0.0252532958984375,
0.01715087890625,
-0.0269012451171875,
0.0222015380859375,
0.0062255859375,
0.042999267578125,
-0.040863037109375,
0.02178955078125,
-0.020721435546875,
0.001224517822265625,
-0.0200042724609375,
0.0103912353515625,
-0.0091094970703125,
-0.036865234375,
0.010772705078125,
-0.07855224609375,
0.0090789794921875,
-0.006801605224609375,
0.084716796875,
0.0102386474609375,
-0.0423583984375,
-0.00559234619140625,
-0.020751953125,
0.039459228515625,
-0.06475830078125,
0.0158233642578125,
0.0318603515625,
0.008453369140625,
-0.0215911865234375,
-0.06500244140625,
-0.043365478515625,
-0.00542449951171875,
-0.0106201171875,
0.0279541015625,
-0.024078369140625,
-0.00023281574249267578,
0.02679443359375,
0.05474853515625,
-0.0645751953125,
-0.006862640380859375,
-0.0258331298828125,
-0.01165008544921875,
0.06170654296875,
0.003940582275390625,
0.025177001953125,
-0.0094451904296875,
-0.0287628173828125,
-0.0438232421875,
-0.05084228515625,
0.0225982666015625,
0.0360107421875,
0.007709503173828125,
-0.0506591796875,
0.042205810546875,
-0.018463134765625,
0.03326416015625,
-0.00218963623046875,
-0.0162811279296875,
0.03143310546875,
-0.040191650390625,
-0.0287628173828125,
-0.021942138671875,
0.0880126953125,
0.031494140625,
-0.01788330078125,
0.0211029052734375,
0.001377105712890625,
0.0024166107177734375,
-0.011383056640625,
-0.062469482421875,
-0.035430908203125,
0.034515380859375,
-0.0423583984375,
-0.01380157470703125,
0.01000213623046875,
-0.060791015625,
-0.00545501708984375,
0.0012493133544921875,
0.036529541015625,
-0.042999267578125,
-0.037139892578125,
0.0223236083984375,
-0.0256500244140625,
0.041656494140625,
0.0262451171875,
-0.046234130859375,
0.0389404296875,
0.0298004150390625,
0.060455322265625,
0.01105499267578125,
-0.01114654541015625,
-0.01015472412109375,
0.005084991455078125,
-0.006610870361328125,
0.0372314453125,
-0.01195526123046875,
-0.033233642578125,
-0.0287628173828125,
0.0230865478515625,
0.0014858245849609375,
-0.019134521484375,
0.04339599609375,
-0.0158233642578125,
0.028167724609375,
-0.04388427734375,
-0.0253143310546875,
-0.035888671875,
-0.0053558349609375,
-0.0567626953125,
0.09136962890625,
0.0305023193359375,
-0.06829833984375,
0.0180206298828125,
-0.050262451171875,
-0.010772705078125,
-0.00270843505859375,
0.0083160400390625,
-0.039093017578125,
-0.010498046875,
0.01739501953125,
0.0129241943359375,
-0.026153564453125,
-0.0017557144165039062,
-0.034027099609375,
-0.00473785400390625,
0.01334381103515625,
-0.01641845703125,
0.1007080078125,
0.015655517578125,
-0.0215911865234375,
-0.004825592041015625,
-0.04205322265625,
-0.0009279251098632812,
0.04400634765625,
-0.016143798828125,
-0.01171875,
-0.01473236083984375,
0.0016336441040039062,
0.00957489013671875,
0.0227508544921875,
-0.038848876953125,
0.039703369140625,
-0.0184173583984375,
0.060791015625,
0.03857421875,
-0.0029926300048828125,
0.0162353515625,
-0.034576416015625,
0.036712646484375,
0.0005998611450195312,
0.046630859375,
0.002117156982421875,
-0.0662841796875,
-0.048095703125,
-0.02740478515625,
0.0284881591796875,
0.04400634765625,
-0.039764404296875,
0.037750244140625,
-0.004375457763671875,
-0.051666259765625,
-0.0298614501953125,
-0.0171356201171875,
0.0256805419921875,
0.0247955322265625,
0.03228759765625,
-0.04583740234375,
-0.0284423828125,
-0.059722900390625,
0.0121002197265625,
-0.033111572265625,
0.00421905517578125,
0.0372314453125,
0.06329345703125,
-0.0006546974182128906,
0.057952880859375,
-0.04522705078125,
-0.0155487060546875,
0.0091705322265625,
0.004390716552734375,
0.0214080810546875,
0.045928955078125,
0.065185546875,
-0.0557861328125,
-0.04034423828125,
-0.005863189697265625,
-0.045745849609375,
-0.0033130645751953125,
0.00507354736328125,
-0.024749755859375,
0.020660400390625,
-0.007312774658203125,
-0.09033203125,
0.0540771484375,
0.04071044921875,
-0.045074462890625,
0.0494384765625,
-0.020843505859375,
0.0036296844482421875,
-0.0667724609375,
0.0164642333984375,
0.0020465850830078125,
-0.0197296142578125,
-0.02447509765625,
0.013702392578125,
-0.001567840576171875,
0.013519287109375,
-0.0279541015625,
0.045867919921875,
-0.035675048828125,
-0.0105133056640625,
0.0128021240234375,
-0.00264739990234375,
0.01715087890625,
0.038482666015625,
-0.004180908203125,
0.05560302734375,
0.040863037109375,
-0.02203369140625,
0.035400390625,
0.040374755859375,
-0.0020618438720703125,
0.0222015380859375,
-0.0634765625,
0.01053619384765625,
0.00868988037109375,
0.04681396484375,
-0.0767822265625,
-0.0181427001953125,
0.042236328125,
-0.04168701171875,
0.0255889892578125,
-0.020233154296875,
-0.039947509765625,
-0.037078857421875,
-0.05096435546875,
0.032135009765625,
0.053558349609375,
-0.041656494140625,
0.029205322265625,
0.033447265625,
-0.0006694793701171875,
-0.047271728515625,
-0.04974365234375,
-0.018951416015625,
-0.0233001708984375,
-0.053314208984375,
0.036651611328125,
-0.005054473876953125,
0.003047943115234375,
0.006683349609375,
-0.0053558349609375,
0.004650115966796875,
-0.010833740234375,
0.03009033203125,
0.02264404296875,
-0.01528167724609375,
-0.015869140625,
0.00630950927734375,
0.0016927719116210938,
0.0006346702575683594,
-0.0286102294921875,
0.037567138671875,
-0.017913818359375,
-0.00036525726318359375,
-0.03546142578125,
0.00881195068359375,
0.039703369140625,
-0.016998291015625,
0.05426025390625,
0.04986572265625,
-0.019073486328125,
0.01457977294921875,
-0.0333251953125,
-0.007556915283203125,
-0.032928466796875,
0.0010013580322265625,
-0.02203369140625,
-0.054412841796875,
0.05621337890625,
0.038299560546875,
0.015869140625,
0.06304931640625,
0.03497314453125,
0.0048065185546875,
0.07080078125,
0.0287322998046875,
-0.00249481201171875,
0.0330810546875,
-0.046875,
-0.01446533203125,
-0.061859130859375,
-0.033050537109375,
-0.031585693359375,
-0.0198822021484375,
-0.059844970703125,
-0.03717041015625,
0.02264404296875,
0.0225982666015625,
-0.046417236328125,
0.049163818359375,
-0.05419921875,
0.01508331298828125,
0.049957275390625,
0.0262603759765625,
0.01541900634765625,
0.0002884864807128906,
-0.0169677734375,
0.0149078369140625,
-0.041900634765625,
-0.0251312255859375,
0.0906982421875,
0.0335693359375,
0.041107177734375,
0.0196990966796875,
0.0265045166015625,
-0.0017137527465820312,
0.02294921875,
-0.032806396484375,
0.03814697265625,
-0.00780487060546875,
-0.051544189453125,
-0.021881103515625,
-0.039093017578125,
-0.0765380859375,
0.02423095703125,
-0.0023193359375,
-0.060791015625,
0.0302581787109375,
-0.002079010009765625,
-0.034271240234375,
0.0260162353515625,
-0.053497314453125,
0.08380126953125,
-0.01312255859375,
-0.0255889892578125,
-0.004520416259765625,
-0.064697265625,
0.033111572265625,
0.01189422607421875,
-0.00328826904296875,
-0.015899658203125,
-0.0313720703125,
0.06622314453125,
-0.0748291015625,
0.053375244140625,
-0.0239410400390625,
-0.00209808349609375,
0.05120849609375,
-0.007747650146484375,
0.036651611328125,
0.0186920166015625,
0.004241943359375,
0.030426025390625,
0.038116455078125,
-0.03399658203125,
-0.0164031982421875,
0.0394287109375,
-0.0823974609375,
-0.01983642578125,
-0.042388916015625,
-0.035003662109375,
0.00862884521484375,
0.0013561248779296875,
0.031402587890625,
0.023651123046875,
-0.0002989768981933594,
0.0093536376953125,
0.04510498046875,
-0.036224365234375,
0.032379150390625,
0.032135009765625,
-0.019317626953125,
-0.05523681640625,
0.05712890625,
0.007476806640625,
0.00988006591796875,
0.016632080078125,
0.01235198974609375,
-0.043701171875,
-0.03558349609375,
-0.038604736328125,
0.03863525390625,
-0.034149169921875,
-0.03436279296875,
-0.0482177734375,
-0.01172637939453125,
-0.0287933349609375,
0.0214996337890625,
-0.0274505615234375,
-0.05389404296875,
-0.04010009765625,
0.0013036727905273438,
0.06842041015625,
0.03594970703125,
-0.0200958251953125,
0.030670166015625,
-0.047760009765625,
0.0167999267578125,
0.02362060546875,
0.0182342529296875,
-0.00881195068359375,
-0.05035400390625,
0.0145111083984375,
0.0135955810546875,
-0.044677734375,
-0.07769775390625,
0.04827880859375,
0.01413726806640625,
0.031524658203125,
0.025238037109375,
0.00909423828125,
0.06396484375,
-0.002346038818359375,
0.07513427734375,
0.0230560302734375,
-0.06439208984375,
0.038238525390625,
-0.03570556640625,
0.00803375244140625,
0.0335693359375,
0.0391845703125,
-0.01377105712890625,
-0.0261383056640625,
-0.06341552734375,
-0.0670166015625,
0.047698974609375,
0.030426025390625,
-0.006931304931640625,
0.01508331298828125,
0.04681396484375,
0.01230621337890625,
0.01337432861328125,
-0.06951904296875,
-0.0389404296875,
-0.034759521484375,
-0.00420379638671875,
0.00543975830078125,
-0.00740814208984375,
-0.01345062255859375,
-0.044921875,
0.0611572265625,
-0.00983428955078125,
0.040496826171875,
0.0232391357421875,
0.01422119140625,
-0.01088714599609375,
-0.007068634033203125,
0.034820556640625,
0.046051025390625,
-0.0261383056640625,
-0.019378662109375,
0.005649566650390625,
-0.05975341796875,
-0.0027313232421875,
0.0262451171875,
-0.00341033935546875,
-0.01360321044921875,
0.007450103759765625,
0.054168701171875,
-0.0140533447265625,
-0.033447265625,
0.0421142578125,
-0.02099609375,
-0.027496337890625,
-0.0186920166015625,
0.006664276123046875,
0.0165252685546875,
0.0280303955078125,
0.0212860107421875,
-0.018707275390625,
0.01053619384765625,
-0.0438232421875,
0.00299835205078125,
0.0433349609375,
0.0008301734924316406,
-0.0341796875,
0.06427001953125,
-0.006496429443359375,
0.01113128662109375,
0.047698974609375,
-0.0206451416015625,
-0.023101806640625,
0.064453125,
0.0310821533203125,
0.046722412109375,
-0.01849365234375,
0.0169677734375,
0.042236328125,
0.01313018798828125,
-0.00969696044921875,
0.04364013671875,
-0.00136566162109375,
-0.035552978515625,
-0.0262451171875,
-0.04541015625,
-0.0206756591796875,
0.0295562744140625,
-0.061126708984375,
0.01552581787109375,
-0.037353515625,
-0.03289794921875,
-0.0033321380615234375,
0.024169921875,
-0.03546142578125,
0.012359619140625,
0.00951385498046875,
0.0826416015625,
-0.0626220703125,
0.06829833984375,
0.047760009765625,
-0.044342041015625,
-0.07244873046875,
-0.0283660888671875,
0.0025272369384765625,
-0.0506591796875,
0.01291656494140625,
0.006587982177734375,
0.02996826171875,
-0.006900787353515625,
-0.058135986328125,
-0.07000732421875,
0.11328125,
0.0328369140625,
-0.034515380859375,
0.0011873245239257812,
0.0013599395751953125,
0.0308380126953125,
-0.0152435302734375,
0.053070068359375,
0.0457763671875,
0.0318603515625,
0.0216522216796875,
-0.08233642578125,
0.0301361083984375,
-0.0180816650390625,
-0.0017414093017578125,
0.01837158203125,
-0.0858154296875,
0.084716796875,
-0.0012140274047851562,
-0.012237548828125,
0.0257720947265625,
0.041900634765625,
0.03363037109375,
0.0036716461181640625,
0.0321044921875,
0.07318115234375,
0.05621337890625,
-0.027252197265625,
0.08306884765625,
-0.01092529296875,
0.04736328125,
0.05914306640625,
0.0016889572143554688,
0.062164306640625,
0.02264404296875,
-0.05096435546875,
0.041168212890625,
0.08221435546875,
-0.006610870361328125,
0.033935546875,
-0.0027599334716796875,
-0.01495361328125,
0.0036640167236328125,
0.00201416015625,
-0.057952880859375,
0.00958251953125,
0.033172607421875,
-0.0193939208984375,
-0.001979827880859375,
-0.0100250244140625,
0.0034008026123046875,
-0.051116943359375,
-0.01678466796875,
0.0419921875,
0.018707275390625,
-0.025146484375,
0.0604248046875,
-0.00217437744140625,
0.045196533203125,
-0.042144775390625,
-0.0024318695068359375,
-0.029449462890625,
-0.0038127899169921875,
-0.0283203125,
-0.06689453125,
0.01418304443359375,
-0.0175933837890625,
0.00522613525390625,
-0.0026760101318359375,
0.0413818359375,
-0.01148223876953125,
-0.0162506103515625,
0.0278472900390625,
0.02337646484375,
0.0262603759765625,
-0.0104522705078125,
-0.0849609375,
0.0221710205078125,
-0.00020503997802734375,
-0.048492431640625,
0.0305023193359375,
0.0287628173828125,
0.00676727294921875,
0.049560546875,
0.0477294921875,
-0.004573822021484375,
0.005970001220703125,
-0.0228118896484375,
0.0758056640625,
-0.04803466796875,
-0.0290374755859375,
-0.0540771484375,
0.0440673828125,
-0.0012388229370117188,
-0.034515380859375,
0.053375244140625,
0.055328369140625,
0.06231689453125,
0.00560760498046875,
0.0450439453125,
-0.036224365234375,
0.01861572265625,
-0.024200439453125,
0.040191650390625,
-0.053466796875,
0.0016870498657226562,
-0.01451873779296875,
-0.062469482421875,
-0.0120849609375,
0.054595947265625,
-0.01541900634765625,
0.00945281982421875,
0.039947509765625,
0.061492919921875,
-0.0046539306640625,
0.0164642333984375,
-0.0004591941833496094,
0.029449462890625,
0.00948333740234375,
0.070556640625,
0.059112548828125,
-0.061370849609375,
0.04510498046875,
-0.035614013671875,
-0.021514892578125,
-0.006801605224609375,
-0.061126708984375,
-0.05401611328125,
-0.0292510986328125,
-0.050872802734375,
-0.04827880859375,
-0.0025539398193359375,
0.058502197265625,
0.056549072265625,
-0.054901123046875,
-0.0298004150390625,
-0.0092620849609375,
-0.0032749176025390625,
-0.0207672119140625,
-0.0194549560546875,
0.034759521484375,
0.0250396728515625,
-0.0374755859375,
0.003192901611328125,
0.009735107421875,
0.0292205810546875,
-0.010711669921875,
-0.02154541015625,
-0.01059722900390625,
-0.0012378692626953125,
0.039581298828125,
0.053680419921875,
-0.051849365234375,
-0.006988525390625,
-0.0182952880859375,
-0.008880615234375,
0.0261077880859375,
0.024078369140625,
-0.056060791015625,
0.00542449951171875,
0.038665771484375,
0.0030841827392578125,
0.06610107421875,
0.007129669189453125,
0.013397216796875,
-0.025146484375,
0.0228424072265625,
0.007183074951171875,
0.0263214111328125,
0.00380706787109375,
-0.0248260498046875,
0.05523681640625,
0.031280517578125,
-0.043212890625,
-0.0595703125,
-0.01468658447265625,
-0.10357666015625,
-0.0198516845703125,
0.0811767578125,
-0.02117919921875,
-0.03363037109375,
0.00563812255859375,
-0.025421142578125,
0.024078369140625,
-0.040374755859375,
0.0179901123046875,
0.024993896484375,
-0.0206298828125,
-0.020050048828125,
-0.06427001953125,
0.036529541015625,
0.0140838623046875,
-0.061676025390625,
-0.00569915771484375,
0.032196044921875,
0.035552978515625,
0.0114593505859375,
0.05230712890625,
-0.028167724609375,
0.024322509765625,
0.0026760101318359375,
0.006374359130859375,
-0.003143310546875,
0.007259368896484375,
-0.026519775390625,
0.0072021484375,
-0.0125885009765625,
-0.0126953125
]
] |
CompVis/stable-diffusion-v1-2 | 2023-07-05T16:18:11.000Z | [
"diffusers",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"arxiv:2112.10752",
"arxiv:2103.00020",
"arxiv:2205.11487",
"arxiv:2207.12598",
"arxiv:1910.09700",
"license:creativeml-openrail-m",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | CompVis | null | null | CompVis/stable-diffusion-v1-2 | 34 | 7,617 | diffusers | 2022-08-19T10:24:37 | ---
license: creativeml-openrail-m
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
inference: false
extra_gated_prompt: |-
This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage.
The CreativeML OpenRAIL License specifies:
1. You can't use the model to deliberately produce nor share illegal or harmful outputs or content
2. The authors claim no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license
3. You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully)
Please read the full license carefully here: https://huggingface.co/spaces/CompVis/stable-diffusion-license
extra_gated_heading: Please read the LICENSE to access this model
---
# Stable Diffusion v1-2 Model Card
Stable Diffusion is a latent text-to-image diffusion model capable of generating photo-realistic images given any text input.
For more information about how Stable Diffusion functions, please have a look at [🤗's Stable Diffusion with D🧨iffusers blog](https://huggingface.co/blog/stable_diffusion).
The **Stable-Diffusion-v1-2** checkpoint was initialized with the weights of the [Stable-Diffusion-v1-1](https:/steps/huggingface.co/CompVis/stable-diffusion-v1-1)
checkpoint and subsequently fine-tuned on 515,000 steps at resolution `512x512` on "laion-improved-aesthetics" (a subset of laion2B-en,
filtered to images with an original size `>= 512x512`, estimated aesthetics score `> 5.0`, and an estimated watermark probability `< 0.5`.
For more information, please refer to [Training](#training).
This weights here are intended to be used with the D🧨iffusers library. If you are looking for the weights to be loaded into the CompVis Stable Diffusion codebase, [come here](https://huggingface.co/CompVis/stable-diffusion-v-1-2-original)
## Model Details
- **Developed by:** Robin Rombach, Patrick Esser
- **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.
- **Model Description:** This is a model that can be used to generate and modify images based on text prompts. It is a [Latent Diffusion Model](https://arxiv.org/abs/2112.10752) that uses a fixed, pretrained text encoder ([CLIP ViT-L/14](https://arxiv.org/abs/2103.00020)) as suggested in the [Imagen paper](https://arxiv.org/abs/2205.11487).
- **Resources for more information:** [GitHub Repository](https://github.com/CompVis/stable-diffusion), [Paper](https://arxiv.org/abs/2112.10752).
- **Cite as:**
@InProceedings{Rombach_2022_CVPR,
author = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
title = {High-Resolution Image Synthesis With Latent Diffusion Models},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2022},
pages = {10684-10695}
}
## Examples
We recommend using [🤗's Diffusers library](https://github.com/huggingface/diffusers) to run Stable Diffusion.
```bash
pip install --upgrade diffusers transformers scipy
```
Running the pipeline with the default PNDM scheduler:
```python
import torch
from torch import autocast
from diffusers import StableDiffusionPipeline
model_id = "CompVis/stable-diffusion-v1-2"
device = "cuda"
pipe = StableDiffusionPipeline.from_pretrained(model_id)
pipe = pipe.to(device)
prompt = "a photo of an astronaut riding a horse on mars"
with autocast("cuda"):
image = pipe(prompt)["sample"][0]
image.save("astronaut_rides_horse.png")
```
**Note**:
If you are limited by GPU memory and have less than 10GB of GPU RAM available, please make sure to load the StableDiffusionPipeline in float16 precision instead of the default float32 precision as done above. You can do so by telling diffusers to expect the weights to be in float16 precision:
```py
import torch
pipe = StableDiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.float16)
pipe = pipe.to(device)
prompt = "a photo of an astronaut riding a horse on mars"
with autocast("cuda"):
image = pipe(prompt, guidance_scale=7.5)["sample"][0]
image.save("astronaut_rides_horse.png")
```
To swap out the noise scheduler, pass it to `from_pretrained`:
```python
from diffusers import StableDiffusionPipeline, LMSDiscreteScheduler
model_id = "CompVis/stable-diffusion-v1-2"
# Use the K-LMS scheduler here instead
scheduler = LMSDiscreteScheduler(beta_start=0.00085, beta_end=0.012, beta_schedule="scaled_linear", num_train_timesteps=1000)
pipe = StableDiffusionPipeline.from_pretrained(model_id, scheduler=scheduler, use_auth_token=True)
pipe = pipe.to("cuda")
prompt = "a photo of an astronaut riding a horse on mars"
with autocast("cuda"):
image = pipe(prompt, guidance_scale=7.5)["sample"][0]
image.save("astronaut_rides_horse.png")
```
# Uses
## Direct Use
The model is intended for research purposes only. 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 generative models.
- Generation of artworks and use in design and other artistic processes.
- Applications in educational or creative tools.
- Research on generative models.
Excluded uses are described below.
### Misuse, Malicious Use, and Out-of-Scope Use
_Note: This section is taken from the [DALLE-MINI model card](https://huggingface.co/dalle-mini/dalle-mini), but applies in the same way to Stable Diffusion v1_.
The model should not be used to intentionally create or disseminate images that create hostile or alienating environments for people. This includes generating images that people would foreseeably find disturbing, distressing, or offensive; or content that propagates historical or current stereotypes.
#### Out-of-Scope Use
The model was not trained to be factual or true representations of people or events, and therefore using the model to generate such content is out-of-scope for the abilities of this model.
#### Misuse and Malicious Use
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 demeaning, dehumanizing, or otherwise harmful representations of people or their environments, cultures, religions, etc.
- Intentionally promoting or propagating discriminatory content or harmful stereotypes.
- Impersonating individuals without their consent.
- Sexual content without consent of the people who might see it.
- Mis- and disinformation
- Representations of egregious violence and gore
- Sharing of copyrighted or licensed material in violation of its terms of use.
- Sharing content that is an alteration of copyrighted or licensed material in violation of its terms of use.
## Limitations and Bias
### Limitations
- The model does not achieve perfect photorealism
- The model cannot render legible text
- The model does not perform well on more difficult tasks which involve compositionality, such as rendering an image corresponding to “A red cube on top of a blue sphere”
- Faces and people in general may not be generated properly.
- The model was trained mainly with English captions and will not work as well in other languages.
- The autoencoding part of the model is lossy
- The model was trained on a large-scale dataset
[LAION-5B](https://laion.ai/blog/laion-5b/) which contains adult material
and is not fit for product use without additional safety mechanisms and
considerations.
- No additional measures were used to deduplicate the dataset. As a result, we observe some degree of memorization for images that are duplicated in the training data.
The training data can be searched at [https://rom1504.github.io/clip-retrieval/](https://rom1504.github.io/clip-retrieval/) to possibly assist in the detection of memorized images.
### Bias
While the capabilities of image generation models are impressive, they can also reinforce or exacerbate social biases.
Stable Diffusion v1 was trained on subsets of [LAION-2B(en)](https://laion.ai/blog/laion-5b/),
which consists of images that are primarily limited to English descriptions.
Texts and images from communities and cultures that use other languages are likely to be insufficiently accounted for.
This affects the overall output of the model, as white and western cultures are often set as the default. Further, the
ability of the model to generate content with non-English prompts is significantly worse than with English-language prompts.
## Training
### Training Data
The model developers used the following dataset for training the model:
- LAION-2B (en) and subsets thereof (see next section)
### Training Procedure
Stable Diffusion v1-4 is a latent diffusion model which combines an autoencoder with a diffusion model that is trained in the latent space of the autoencoder. During training,
- Images are encoded through an encoder, which turns images into latent representations. The autoencoder uses a relative downsampling factor of 8 and maps images of shape H x W x 3 to latents of shape H/f x W/f x 4
- Text prompts are encoded through a ViT-L/14 text-encoder.
- The non-pooled output of the text encoder is fed into the UNet backbone of the latent diffusion model via cross-attention.
- The loss is a reconstruction objective between the noise that was added to the latent and the prediction made by the UNet.
We currently provide four checkpoints, which were trained as follows.
- [`stable-diffusion-v1-1`](https://huggingface.co/CompVis/stable-diffusion-v1-1): 237,000 steps at resolution `256x256` on [laion2B-en](https://huggingface.co/datasets/laion/laion2B-en).
194,000 steps at resolution `512x512` on [laion-high-resolution](https://huggingface.co/datasets/laion/laion-high-resolution) (170M examples from LAION-5B with resolution `>= 1024x1024`).
- [`stable-diffusion-v1-2`](https://huggingface.co/CompVis/stable-diffusion-v1-2): Resumed from `stable-diffusion-v1-1`.
515,000 steps at resolution `512x512` on "laion-improved-aesthetics" (a subset of laion2B-en,
filtered to images with an original size `>= 512x512`, estimated aesthetics score `> 5.0`, and an estimated watermark probability `< 0.5`. The watermark estimate is from the LAION-5B metadata, the aesthetics score is estimated using an [improved aesthetics estimator](https://github.com/christophschuhmann/improved-aesthetic-predictor)).
- [`stable-diffusion-v1-3`](https://huggingface.co/CompVis/stable-diffusion-v1-3): Resumed from `stable-diffusion-v1-2`. 195,000 steps at resolution `512x512` on "laion-improved-aesthetics" and 10 % dropping of the text-conditioning to improve [classifier-free guidance sampling](https://arxiv.org/abs/2207.12598).
- [**`stable-diffusion-v1-4`**](https://huggingface.co/CompVis/stable-diffusion-v1-4) Resumed from `stable-diffusion-v1-2`.225,000 steps at resolution `512x512` on "laion-aesthetics v2 5+" and 10 % dropping of the text-conditioning to improve [classifier-free guidance sampling](https://arxiv.org/abs/2207.12598).
### Training details
- **Hardware:** 32 x 8 x A100 GPUs
- **Optimizer:** AdamW
- **Gradient Accumulations**: 2
- **Batch:** 32 x 8 x 2 x 4 = 2048
- **Learning rate:** warmup to 0.0001 for 10,000 steps and then kept constant
## Evaluation Results
Evaluations with different classifier-free guidance scales (1.5, 2.0, 3.0, 4.0,
5.0, 6.0, 7.0, 8.0) and 50 PLMS sampling
steps show the relative improvements of the checkpoints:
Evaluated using 50 PLMS steps and 10000 random prompts from the COCO2017 validation set, evaluated at 512x512 resolution. Not optimized for FID scores.
## Environmental Impact
**Stable Diffusion v1** **Estimated Emissions**
Based on that information, we estimate the following CO2 emissions using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). The hardware, runtime, cloud provider, and compute region were utilized to estimate the carbon impact.
- **Hardware Type:** A100 PCIe 40GB
- **Hours used:** 150000
- **Cloud Provider:** AWS
- **Compute Region:** US-east
- **Carbon Emitted (Power consumption x Time x Carbon produced based on location of power grid):** 11250 kg CO2 eq.
## Citation
```bibtex
@InProceedings{Rombach_2022_CVPR,
author = {Rombach, Robin and Blattmann, Andreas and Lorenz, Dominik and Esser, Patrick and Ommer, Bj\"orn},
title = {High-Resolution Image Synthesis With Latent Diffusion Models},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2022},
pages = {10684-10695}
}
```
*This model card was written by: Robin Rombach and Patrick Esser and is based on the [DALL-E Mini model card](https://huggingface.co/dalle-mini/dalle-mini).* | 13,995 | [
[
-0.0276336669921875,
-0.061004638671875,
0.0281219482421875,
0.01806640625,
-0.0216522216796875,
-0.0274505615234375,
0.0018367767333984375,
-0.0256195068359375,
-0.0036163330078125,
0.0321044921875,
-0.0237579345703125,
-0.03765869140625,
-0.0550537109375,
-0.0164337158203125,
-0.0296173095703125,
0.065185546875,
-0.01354217529296875,
-0.00014925003051757812,
-0.0101470947265625,
-0.0065460205078125,
-0.021087646484375,
-0.01922607421875,
-0.069091796875,
-0.0191802978515625,
0.042388916015625,
0.005970001220703125,
0.046966552734375,
0.04754638671875,
0.0380859375,
0.0224151611328125,
-0.0182952880859375,
0.0029735565185546875,
-0.039276123046875,
-0.004787445068359375,
0.0009379386901855469,
-0.0221710205078125,
-0.0382080078125,
0.013092041015625,
0.052459716796875,
0.0200042724609375,
-0.0102081298828125,
0.0015764236450195312,
0.0005526542663574219,
0.04632568359375,
-0.04193115234375,
-0.0006856918334960938,
-0.0250701904296875,
0.01059722900390625,
-0.0080718994140625,
0.006427764892578125,
-0.0278167724609375,
-0.014404296875,
0.005527496337890625,
-0.06494140625,
0.026641845703125,
-0.0174407958984375,
0.08856201171875,
0.0271148681640625,
-0.019927978515625,
-0.0174560546875,
-0.053314208984375,
0.0499267578125,
-0.048431396484375,
0.01515960693359375,
0.02447509765625,
0.0085296630859375,
0.0016145706176757812,
-0.08648681640625,
-0.049072265625,
-0.0118560791015625,
-0.0006694793701171875,
0.03399658203125,
-0.0159149169921875,
-0.0032806396484375,
0.02880859375,
0.017608642578125,
-0.043548583984375,
-0.0040130615234375,
-0.04248046875,
-0.004871368408203125,
0.048187255859375,
0.007904052734375,
0.024688720703125,
-0.0107879638671875,
-0.033966064453125,
-0.0144805908203125,
-0.03729248046875,
-0.00250244140625,
0.027008056640625,
-0.019500732421875,
-0.029449462890625,
0.0369873046875,
0.0167694091796875,
0.0384521484375,
0.0228729248046875,
-0.00907135009765625,
0.0298614501953125,
-0.0179443359375,
-0.0183868408203125,
-0.02880859375,
0.07183837890625,
0.04156494140625,
-0.0014448165893554688,
0.00887298583984375,
-0.01161956787109375,
0.0056304931640625,
0.0038700103759765625,
-0.0882568359375,
-0.04278564453125,
0.0107269287109375,
-0.05059814453125,
-0.0377197265625,
-0.01580810546875,
-0.06488037109375,
-0.0187835693359375,
0.004730224609375,
0.050750732421875,
-0.033111572265625,
-0.033721923828125,
0.006954193115234375,
-0.0270233154296875,
0.0076751708984375,
0.033538818359375,
-0.048797607421875,
0.001720428466796875,
-0.001522064208984375,
0.08367919921875,
-0.0200347900390625,
-0.01334381103515625,
-0.005702972412109375,
0.01242828369140625,
-0.0219879150390625,
0.050811767578125,
-0.016143798828125,
-0.0478515625,
-0.02294921875,
0.0283050537109375,
0.0018262863159179688,
-0.045166015625,
0.041412353515625,
-0.03936767578125,
0.021484375,
-0.004985809326171875,
-0.0400390625,
-0.01494598388671875,
-0.0010585784912109375,
-0.051971435546875,
0.076171875,
0.016754150390625,
-0.06689453125,
0.0182647705078125,
-0.054656982421875,
-0.01544189453125,
-0.0033168792724609375,
0.000025391578674316406,
-0.057891845703125,
-0.0211181640625,
0.0011072158813476562,
0.0283050537109375,
-0.0033740997314453125,
0.01451873779296875,
-0.0189208984375,
-0.014617919921875,
-0.0024433135986328125,
-0.0372314453125,
0.08013916015625,
0.0303497314453125,
-0.0296478271484375,
0.004756927490234375,
-0.04937744140625,
-0.0179901123046875,
0.0318603515625,
-0.02398681640625,
-0.01397705078125,
-0.011474609375,
0.0234222412109375,
0.022003173828125,
0.01290130615234375,
-0.0305328369140625,
0.007228851318359375,
-0.020263671875,
0.037933349609375,
0.05853271484375,
0.02093505859375,
0.048858642578125,
-0.0357666015625,
0.0382080078125,
0.02899169921875,
0.0164031982421875,
-0.0182952880859375,
-0.05584716796875,
-0.0560302734375,
-0.0338134765625,
0.012054443359375,
0.04168701171875,
-0.0552978515625,
0.02789306640625,
-0.00012046098709106445,
-0.057159423828125,
-0.016845703125,
-0.0008854866027832031,
0.016357421875,
0.0546875,
0.02801513671875,
-0.033447265625,
-0.0238494873046875,
-0.057708740234375,
0.0171051025390625,
-0.006046295166015625,
0.008514404296875,
0.0263519287109375,
0.05279541015625,
-0.0222625732421875,
0.044036865234375,
-0.041839599609375,
-0.0270843505859375,
0.01305389404296875,
0.0200653076171875,
0.008575439453125,
0.06207275390625,
0.05804443359375,
-0.0665283203125,
-0.052032470703125,
-0.01715087890625,
-0.0667724609375,
0.00222015380859375,
-0.0102996826171875,
-0.02410888671875,
0.022247314453125,
0.04052734375,
-0.06488037109375,
0.045623779296875,
0.03424072265625,
-0.0260162353515625,
0.041229248046875,
-0.036773681640625,
0.00926971435546875,
-0.07666015625,
0.0181732177734375,
0.0169830322265625,
-0.0251007080078125,
-0.03607177734375,
0.00424957275390625,
-0.005290985107421875,
-0.00806427001953125,
-0.05035400390625,
0.059722900390625,
-0.033294677734375,
0.0328369140625,
-0.023712158203125,
-0.0049591064453125,
0.0116119384765625,
0.0236053466796875,
0.0239410400390625,
0.048065185546875,
0.06964111328125,
-0.0509033203125,
0.0206146240234375,
0.0181732177734375,
-0.0189971923828125,
0.044281005859375,
-0.0693359375,
0.01116180419921875,
-0.0306854248046875,
0.022216796875,
-0.0784912109375,
-0.01074981689453125,
0.046356201171875,
-0.0303955078125,
0.031646728515625,
-0.0173797607421875,
-0.038726806640625,
-0.029876708984375,
-0.005615234375,
0.044189453125,
0.07257080078125,
-0.02886962890625,
0.0367431640625,
0.0285797119140625,
0.01030731201171875,
-0.0295562744140625,
-0.062347412109375,
-0.00916290283203125,
-0.0322265625,
-0.058197021484375,
0.041778564453125,
-0.02203369140625,
-0.006038665771484375,
0.01149749755859375,
0.019622802734375,
-0.00826263427734375,
-0.01090240478515625,
0.028472900390625,
0.0206146240234375,
-0.0021514892578125,
-0.005218505859375,
0.0036182403564453125,
-0.007450103759765625,
-0.0015163421630859375,
-0.0166473388671875,
0.025177001953125,
0.008575439453125,
-0.01430511474609375,
-0.0552978515625,
0.0223388671875,
0.043060302734375,
0.00841522216796875,
0.0628662109375,
0.0745849609375,
-0.03668212890625,
-0.0004565715789794922,
-0.0282745361328125,
-0.016632080078125,
-0.038726806640625,
0.0262603759765625,
-0.009674072265625,
-0.03497314453125,
0.050628662109375,
0.002216339111328125,
0.002483367919921875,
0.050506591796875,
0.06097412109375,
-0.0221710205078125,
0.0784912109375,
0.035003662109375,
0.0267333984375,
0.054473876953125,
-0.0640869140625,
-0.008087158203125,
-0.06024169921875,
-0.0233154296875,
-0.0131378173828125,
-0.0256805419921875,
-0.0271148681640625,
-0.053741455078125,
0.032318115234375,
0.01508331298828125,
-0.0250244140625,
0.01045989990234375,
-0.04925537109375,
0.023345947265625,
0.0276031494140625,
0.02130126953125,
0.00452423095703125,
0.01169586181640625,
-0.004505157470703125,
-0.01270294189453125,
-0.055084228515625,
-0.047515869140625,
0.0787353515625,
0.032745361328125,
0.07415771484375,
-0.008056640625,
0.043548583984375,
0.0325927734375,
0.031280517578125,
-0.035552978515625,
0.04058837890625,
-0.0162506103515625,
-0.055755615234375,
-0.01165771484375,
-0.0185699462890625,
-0.07513427734375,
0.013824462890625,
-0.016265869140625,
-0.031829833984375,
0.03564453125,
0.0254974365234375,
-0.018096923828125,
0.033905029296875,
-0.059844970703125,
0.072509765625,
-0.0105133056640625,
-0.0523681640625,
-0.00832366943359375,
-0.04632568359375,
0.02593994140625,
0.0010690689086914062,
0.01184844970703125,
0.0013532638549804688,
-0.006439208984375,
0.06732177734375,
-0.0267181396484375,
0.066650390625,
-0.034637451171875,
0.0048065185546875,
0.041290283203125,
-0.005950927734375,
0.03167724609375,
0.007389068603515625,
-0.00614166259765625,
0.022918701171875,
0.0132904052734375,
-0.03411865234375,
-0.0220794677734375,
0.054168701171875,
-0.074462890625,
-0.025634765625,
-0.032745361328125,
-0.032684326171875,
0.0394287109375,
0.020294189453125,
0.061859130859375,
0.0203094482421875,
-0.01058197021484375,
-0.0016775131225585938,
0.059844970703125,
-0.02899169921875,
0.03411865234375,
0.0185394287109375,
-0.024566650390625,
-0.0408935546875,
0.06732177734375,
0.018829345703125,
0.036590576171875,
-0.00890350341796875,
0.007568359375,
-0.0180206298828125,
-0.038726806640625,
-0.039886474609375,
0.0259857177734375,
-0.068115234375,
-0.00968170166015625,
-0.05731201171875,
-0.0295867919921875,
-0.024505615234375,
-0.01010894775390625,
-0.030914306640625,
-0.026885986328125,
-0.0693359375,
0.0145111083984375,
0.022247314453125,
0.04058837890625,
-0.02496337890625,
0.0362548828125,
-0.035247802734375,
0.028167724609375,
0.01531982421875,
0.01776123046875,
0.004596710205078125,
-0.061309814453125,
-0.01971435546875,
0.00799560546875,
-0.05352783203125,
-0.06488037109375,
0.036041259765625,
0.0102081298828125,
0.044464111328125,
0.03253173828125,
-0.01251983642578125,
0.044830322265625,
-0.03515625,
0.07391357421875,
0.0250701904296875,
-0.044464111328125,
0.044769287109375,
-0.032745361328125,
0.007183074951171875,
0.01285552978515625,
0.03729248046875,
-0.0207977294921875,
-0.02557373046875,
-0.06402587890625,
-0.07049560546875,
0.043060302734375,
0.030792236328125,
0.0188751220703125,
-0.00003451108932495117,
0.051849365234375,
-0.00653839111328125,
-0.0005850791931152344,
-0.08099365234375,
-0.041229248046875,
-0.028411865234375,
-0.00004458427429199219,
0.002410888671875,
-0.0127716064453125,
-0.01221466064453125,
-0.042633056640625,
0.068115234375,
0.0137786865234375,
0.041290283203125,
0.0364990234375,
-0.0027065277099609375,
-0.0282745361328125,
-0.0262603759765625,
0.040069580078125,
0.03424072265625,
-0.0152435302734375,
-0.00021183490753173828,
-0.00830841064453125,
-0.04193115234375,
0.020721435546875,
0.009063720703125,
-0.042938232421875,
0.0061187744140625,
0.0016450881958007812,
0.0594482421875,
-0.0254058837890625,
-0.0316162109375,
0.044189453125,
-0.0103607177734375,
-0.029571533203125,
-0.039276123046875,
0.0093231201171875,
0.01262664794921875,
0.021331787109375,
0.01374053955078125,
0.0328369140625,
0.00627899169921875,
-0.0207672119140625,
0.00963592529296875,
0.043609619140625,
-0.0204925537109375,
-0.0222625732421875,
0.08416748046875,
0.001598358154296875,
-0.018035888671875,
0.0389404296875,
-0.032379150390625,
-0.0149383544921875,
0.059295654296875,
0.04559326171875,
0.059417724609375,
-0.01430511474609375,
0.0267333984375,
0.05389404296875,
0.022247314453125,
-0.0247650146484375,
0.0141143798828125,
0.018524169921875,
-0.053253173828125,
-0.00423431396484375,
-0.0297088623046875,
0.0004715919494628906,
0.01546478271484375,
-0.030242919921875,
0.04083251953125,
-0.039093017578125,
-0.035675048828125,
0.0006136894226074219,
-0.023101806640625,
-0.048492431640625,
0.015655517578125,
0.024993896484375,
0.060150146484375,
-0.076416015625,
0.0615234375,
0.05255126953125,
-0.055389404296875,
-0.045318603515625,
-0.00577545166015625,
-0.017791748046875,
-0.032684326171875,
0.050079345703125,
-0.0031719207763671875,
0.004825592041015625,
0.01438140869140625,
-0.066650390625,
-0.058990478515625,
0.0982666015625,
0.0284576416015625,
-0.01910400390625,
-0.0054779052734375,
-0.0155487060546875,
0.04461669921875,
-0.033355712890625,
0.0226287841796875,
0.0153961181640625,
0.031768798828125,
0.03369140625,
-0.04290771484375,
0.00435638427734375,
-0.02264404296875,
0.0150146484375,
-0.016265869140625,
-0.066162109375,
0.081787109375,
-0.02569580078125,
-0.0230712890625,
0.0231475830078125,
0.046844482421875,
0.02294921875,
0.0246429443359375,
0.034912109375,
0.06640625,
0.044189453125,
-0.003322601318359375,
0.0740966796875,
-0.01001739501953125,
0.04339599609375,
0.056549072265625,
0.0015668869018554688,
0.044036865234375,
0.035400390625,
-0.0147705078125,
0.0458984375,
0.056243896484375,
-0.0259857177734375,
0.059173583984375,
0.00823211669921875,
-0.0196075439453125,
-0.0079193115234375,
0.005458831787109375,
-0.04217529296875,
-0.0015020370483398438,
0.0271148681640625,
-0.05279541015625,
-0.0104827880859375,
0.016571044921875,
-0.005466461181640625,
-0.0174560546875,
-0.0104217529296875,
0.035552978515625,
0.0074920654296875,
-0.03338623046875,
0.059844970703125,
0.00595855712890625,
0.063232421875,
-0.03497314453125,
-0.009429931640625,
-0.00240325927734375,
0.0134735107421875,
-0.0185699462890625,
-0.061248779296875,
0.02899169921875,
-0.01548004150390625,
-0.01470947265625,
-0.0189666748046875,
0.055389404296875,
-0.032989501953125,
-0.0499267578125,
0.01922607421875,
0.020843505859375,
0.0297088623046875,
0.01009368896484375,
-0.0811767578125,
0.01334381103515625,
-0.009552001953125,
-0.031341552734375,
0.0142364501953125,
0.0167236328125,
0.0191497802734375,
0.050537109375,
0.0377197265625,
0.006374359130859375,
0.014892578125,
-0.01230621337890625,
0.061431884765625,
-0.0263519287109375,
-0.0290679931640625,
-0.056915283203125,
0.058746337890625,
-0.00135040283203125,
-0.01380157470703125,
0.05035400390625,
0.0518798828125,
0.04833984375,
-0.0199432373046875,
0.0650634765625,
-0.0232696533203125,
0.006072998046875,
-0.036773681640625,
0.0721435546875,
-0.06396484375,
0.01123046875,
-0.0288543701171875,
-0.059112548828125,
-0.01605224609375,
0.07098388671875,
-0.0196075439453125,
0.0240936279296875,
0.037506103515625,
0.08062744140625,
-0.01904296875,
-0.02899169921875,
0.0208740234375,
0.022247314453125,
0.0288848876953125,
0.0269317626953125,
0.061431884765625,
-0.056610107421875,
0.037689208984375,
-0.03448486328125,
-0.0140533447265625,
-0.00021970272064208984,
-0.0692138671875,
-0.0606689453125,
-0.055694580078125,
-0.06256103515625,
-0.05877685546875,
-0.006343841552734375,
0.036773681640625,
0.078125,
-0.038116455078125,
-0.006587982177734375,
-0.0228424072265625,
0.0014619827270507812,
0.0005054473876953125,
-0.0208587646484375,
0.0328369140625,
0.0012683868408203125,
-0.06591796875,
-0.012054443359375,
0.014862060546875,
0.044189453125,
-0.033538818359375,
-0.0181732177734375,
-0.0234832763671875,
-0.006412506103515625,
0.032196044921875,
0.0210723876953125,
-0.046966552734375,
-0.00010401010513305664,
-0.01232147216796875,
-0.0152435302734375,
0.0149078369140625,
0.0264892578125,
-0.048583984375,
0.039276123046875,
0.037841796875,
0.00882720947265625,
0.06719970703125,
-0.005977630615234375,
0.0052337646484375,
-0.037506103515625,
0.031768798828125,
0.0156402587890625,
0.02557373046875,
0.022705078125,
-0.037445068359375,
0.031036376953125,
0.050018310546875,
-0.06591796875,
-0.056243896484375,
0.006961822509765625,
-0.08648681640625,
-0.01849365234375,
0.0950927734375,
-0.023529052734375,
-0.022369384765625,
0.00902557373046875,
-0.030181884765625,
0.0244903564453125,
-0.0183868408203125,
0.04864501953125,
0.044036865234375,
-0.01263427734375,
-0.035675048828125,
-0.04791259765625,
0.03924560546875,
0.0064849853515625,
-0.049560546875,
-0.018951416015625,
0.048431396484375,
0.05633544921875,
0.0214080810546875,
0.07354736328125,
-0.0226287841796875,
0.0196380615234375,
0.0037994384765625,
0.012481689453125,
0.007732391357421875,
-0.007678985595703125,
-0.026947021484375,
-0.001323699951171875,
-0.0031375885009765625,
-0.01430511474609375
]
] |
jondurbin/airoboros-l2-7b-2.2.1 | 2023-09-21T18:39:31.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:jondurbin/airoboros-2.2.1",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | jondurbin | null | null | jondurbin/airoboros-l2-7b-2.2.1 | 2 | 7,607 | transformers | 2023-09-20T17:26:33 | ---
license: llama2
datasets:
- jondurbin/airoboros-2.2.1
---
### Overview
Another experimental model, using mostly sythetic data generated by [airoboros](https://github.com/jondurbin/airoboros)
This is essentially a minor "fix" branch of [airoboros-l2-7b-2.2](https://hf.co/jondurbin/airoboros-l2-7b-2.2) with a updates, primarily:
- [re-generated writing responses](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#re-generated-writing-responses)
- [longer contextual blocks](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#longer-contextual-blocks)
- [removal of "rp" data](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#rp-category-removed)
This is a fairly general purpose model, but focuses heavily on instruction following, rather than casual chat/roleplay.
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 prompt format:
```
A chat.
USER: {prompt}
ASSISTANT:
```
The default system prompt ("A chat.") was used for most of the prompts, however it also included a wide sampling of responses with other prompts, particularly in "stylized\_response", "rp", "gtkm", etc.
Here's another example:
```
A chat between Bob (aka USER) and Tom (aka ASSISTANT). Tom is an extremely intelligent 18th century bookkeeper, who speaks loquaciously.
USER: {prompt}
ASSISTANT:
```
And chat scenario that wouldn't require USER/ASSISTANT (but should use stopping criteria to prevent the model from speaking on your behalf).
```
A chat between old friends: Timmy and Tommy.
{description of characters}
{setting for the chat}
Timmy: *takes a big sip from his coffee* "Ah, sweet, delicious, magical coffee."
Tommy:
```
__*I strongly suggest adding stopping criteria/early inference stopping on "USER:", and/or whatever names you specify in the system prompt.*__
### Fine tuning info
https://wandb.ai/jondurbin/airoboros-l2-7b-2.2.1/runs/ka6jlcj7?workspace=user-jondurbin
### 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
```
#### Summarization
500 samples have been included from [this dataset](https://huggingface.co/datasets/mattpscott/airoboros-summarization), using the same format as contextual question answering, for example:
```
BEGININPUT
{text to summarize}
ENDINPUT
BEGININSTRUCTION
Summarize the input in around 130 words.
ENDINSTRUCTION
```
#### Getting longer responses
You can use a few techniques to get longer responses.
Detailed prompts, with explicit instruction for word count:
```
Please compose a narrative set in the heart of an ancient library, steeped in the scent of old parchment and ink. The protagonist should be a young scholar who is dedicated to studying the art of storytelling and its evolution throughout history. In her pursuit of knowledge, she stumbles upon a forgotten tome that seems to possess an unusual aura. This book has the ability to bring stories to life, literally manifesting characters and scenarios from within its pages into reality.
The main character must navigate through various epochs of storytelling - from oral traditions of tribal societies, through medieval minstrels' tales, to modern-day digital narratives - as they come alive around her. Each era presents its unique challenges and lessons about the power and impact of stories on human civilization.
One such character could be a sentient quill pen, who was once used by renowned authors of yesteryears and now holds their wisdom and experiences. It becomes her mentor, guiding her through this journey with witty remarks and insightful commentary.
Ensure that your tale encapsulates the thrill of adventure, the beauty of learning, and the profound connection between humans and their stories. All characters involved should be non-human entities. Feel free to explore creative liberties but maintain the mentioned elements.
Your response should be approximately 2300 words.
```
Or, a simpler example:
```
Please create a long, detailed story about a dragon in an old growth forest who, for some reason, begins speaking the words of the source code of linux.
```
#### 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 2.2 models are built on top of llama-2/codellama.
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 mostly 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.
| 18,657 | [
[
-0.0228424072265625,
-0.07952880859375,
0.03485107421875,
0.00634002685546875,
-0.0021305084228515625,
-0.0242156982421875,
-0.01152801513671875,
-0.031951904296875,
0.0269622802734375,
0.039642333984375,
-0.060638427734375,
-0.030487060546875,
-0.023773193359375,
0.0151824951171875,
-0.0149383544921875,
0.08428955078125,
0.0034961700439453125,
-0.0159759521484375,
-0.0162353515625,
0.0012531280517578125,
-0.05322265625,
-0.034027099609375,
-0.0576171875,
-0.0042266845703125,
0.038726806640625,
0.036041259765625,
0.042144775390625,
0.045257568359375,
0.0228424072265625,
0.0259246826171875,
-0.01078033447265625,
0.0229339599609375,
-0.034271240234375,
0.0230255126953125,
-0.0104522705078125,
-0.0352783203125,
-0.032196044921875,
-0.0002415180206298828,
0.03369140625,
0.052276611328125,
-0.0059967041015625,
0.01837158203125,
0.002887725830078125,
0.03240966796875,
-0.03790283203125,
0.00963592529296875,
-0.01343536376953125,
0.001071929931640625,
-0.00916290283203125,
-0.022979736328125,
-0.03765869140625,
-0.01995849609375,
0.0098419189453125,
-0.0687255859375,
-0.00019991397857666016,
0.0256500244140625,
0.06182861328125,
0.00750732421875,
-0.031097412109375,
-0.023468017578125,
-0.042938232421875,
0.05633544921875,
-0.05609130859375,
0.007381439208984375,
0.04510498046875,
0.032989501953125,
-0.0282440185546875,
-0.06597900390625,
-0.048583984375,
-0.0157623291015625,
-0.0209503173828125,
0.01419830322265625,
-0.014862060546875,
-0.00757598876953125,
0.038787841796875,
0.005725860595703125,
-0.057952880859375,
-0.0204315185546875,
-0.044769287109375,
-0.0131072998046875,
0.0389404296875,
0.034759521484375,
0.031402587890625,
-0.03009033203125,
-0.0258941650390625,
-0.01154327392578125,
-0.035186767578125,
0.0299072265625,
0.0225830078125,
0.0290374755859375,
-0.021728515625,
0.045989990234375,
-0.0233917236328125,
0.053070068359375,
0.0024890899658203125,
-0.0154266357421875,
-0.0010957717895507812,
-0.033843994140625,
-0.016876220703125,
-0.01125335693359375,
0.06536865234375,
0.056488037109375,
0.01551055908203125,
0.005435943603515625,
-0.002414703369140625,
-0.01122283935546875,
0.01502227783203125,
-0.06103515625,
-0.0204620361328125,
0.037445068359375,
-0.055023193359375,
-0.0189361572265625,
-0.00841522216796875,
-0.060333251953125,
-0.0280303955078125,
-0.0109100341796875,
0.028045654296875,
-0.039093017578125,
-0.00907135009765625,
0.00939178466796875,
-0.02996826171875,
0.00531768798828125,
0.0390625,
-0.0736083984375,
0.037353515625,
0.023101806640625,
0.060028076171875,
0.00848388671875,
-0.024749755859375,
-0.0300445556640625,
0.00269317626953125,
-0.00937652587890625,
0.06256103515625,
-0.0377197265625,
-0.0312042236328125,
-0.016998291015625,
0.02423095703125,
-0.0014324188232421875,
-0.0242919921875,
0.033233642578125,
-0.0236053466796875,
0.049072265625,
-0.033447265625,
-0.03662109375,
-0.022552490234375,
0.01617431640625,
-0.029205322265625,
0.06951904296875,
0.00884246826171875,
-0.06573486328125,
-0.011932373046875,
-0.06744384765625,
-0.020233154296875,
-0.0029888153076171875,
-0.004680633544921875,
0.01146697998046875,
-0.012969970703125,
0.0099029541015625,
0.034515380859375,
-0.02838134765625,
0.0169830322265625,
-0.0264129638671875,
-0.024322509765625,
0.0316162109375,
-0.0236358642578125,
0.0897216796875,
0.0189208984375,
-0.00917816162109375,
-0.0003330707550048828,
-0.053558349609375,
-0.00347900390625,
0.01207733154296875,
-0.02044677734375,
-0.016876220703125,
-0.00571441650390625,
0.00829315185546875,
-0.004375457763671875,
0.021209716796875,
-0.05133056640625,
0.0330810546875,
-0.02947998046875,
0.049560546875,
0.045318603515625,
0.01299285888671875,
0.0240020751953125,
-0.04486083984375,
0.03948974609375,
-0.005550384521484375,
0.0139617919921875,
-0.05084228515625,
-0.04058837890625,
-0.048553466796875,
0.0037288665771484375,
0.00852203369140625,
0.06610107421875,
-0.04864501953125,
0.027618408203125,
0.0002586841583251953,
-0.0361328125,
-0.019195556640625,
-0.0166015625,
0.0242919921875,
0.06536865234375,
0.038055419921875,
-0.0009551048278808594,
-0.044158935546875,
-0.05889892578125,
0.01001739501953125,
-0.033050537109375,
-0.00684356689453125,
0.040863037109375,
0.03936767578125,
-0.009765625,
0.07745361328125,
-0.07171630859375,
0.005046844482421875,
-0.01291656494140625,
-0.0011348724365234375,
0.013885498046875,
0.05621337890625,
0.034088134765625,
-0.05377197265625,
-0.01953125,
-0.01371002197265625,
-0.06427001953125,
-0.01049041748046875,
-0.01824951171875,
-0.0182037353515625,
0.003971099853515625,
0.03497314453125,
-0.05194091796875,
0.031707763671875,
0.01346588134765625,
-0.048187255859375,
0.0513916015625,
-0.00792694091796875,
0.0246734619140625,
-0.10150146484375,
0.0279693603515625,
-0.01418304443359375,
-0.007602691650390625,
-0.046630859375,
0.0121307373046875,
-0.01226806640625,
-0.014190673828125,
-0.0303802490234375,
0.07049560546875,
-0.028045654296875,
0.01262664794921875,
-0.00405120849609375,
0.011077880859375,
0.0198516845703125,
0.047271728515625,
0.006412506103515625,
0.0477294921875,
0.03607177734375,
-0.042938232421875,
0.046875,
0.031280517578125,
0.00586700439453125,
0.0443115234375,
-0.071044921875,
0.0114288330078125,
-0.01276397705078125,
0.0302886962890625,
-0.08648681640625,
-0.0216522216796875,
0.04644775390625,
-0.053619384765625,
0.0142364501953125,
-0.0018558502197265625,
-0.029083251953125,
-0.02630615234375,
-0.02313232421875,
0.012298583984375,
0.0333251953125,
-0.0228271484375,
0.056121826171875,
0.0159149169921875,
-0.01611328125,
-0.03985595703125,
-0.0545654296875,
0.0128326416015625,
-0.0230255126953125,
-0.054595947265625,
0.0261688232421875,
-0.0406494140625,
-0.0236663818359375,
-0.006969451904296875,
0.0086822509765625,
-0.01751708984375,
0.01544189453125,
0.0240936279296875,
0.0224761962890625,
-0.01522064208984375,
0.007778167724609375,
0.01335906982421875,
0.00009554624557495117,
0.00577545166015625,
-0.01364898681640625,
0.041015625,
-0.0184478759765625,
-0.0118255615234375,
-0.040435791015625,
0.04742431640625,
0.031341552734375,
-0.0011749267578125,
0.044952392578125,
0.038909912109375,
-0.037994384765625,
0.009979248046875,
-0.0296173095703125,
-0.034423828125,
-0.0379638671875,
0.01317596435546875,
-0.029632568359375,
-0.049774169921875,
0.05072021484375,
0.0168304443359375,
0.02215576171875,
0.023956298828125,
0.0227508544921875,
-0.0289764404296875,
0.054168701171875,
0.03668212890625,
0.0199127197265625,
0.024139404296875,
-0.0214691162109375,
-0.0006618499755859375,
-0.06427001953125,
-0.0294647216796875,
-0.0390625,
-0.0253753662109375,
-0.03509521484375,
-0.013427734375,
0.0207977294921875,
0.022857666015625,
-0.021392822265625,
0.03460693359375,
-0.0439453125,
0.0303192138671875,
0.050201416015625,
0.01006317138671875,
0.01103973388671875,
-0.0233917236328125,
-0.0032405853271484375,
-0.0006971359252929688,
-0.054412841796875,
-0.05755615234375,
0.06817626953125,
0.032073974609375,
0.050567626953125,
0.01470184326171875,
0.055389404296875,
0.0152435302734375,
-0.0016736984252929688,
-0.04632568359375,
0.052978515625,
0.00274658203125,
-0.045501708984375,
-0.0322265625,
-0.0159149169921875,
-0.07781982421875,
0.0085296630859375,
-0.01334381103515625,
-0.0787353515625,
0.0213470458984375,
0.014007568359375,
-0.0714111328125,
-0.0004887580871582031,
-0.065185546875,
0.07818603515625,
-0.00547027587890625,
-0.0229339599609375,
0.0147247314453125,
-0.0767822265625,
0.01776123046875,
0.031707763671875,
0.0037364959716796875,
0.006336212158203125,
-0.01015472412109375,
0.0626220703125,
-0.0361328125,
0.0758056640625,
-0.0143280029296875,
0.0179290771484375,
0.036285400390625,
0.01093292236328125,
0.0205078125,
0.016876220703125,
0.0055694580078125,
-0.006771087646484375,
0.0394287109375,
-0.0252227783203125,
-0.052215576171875,
0.049407958984375,
-0.05841064453125,
-0.035003662109375,
-0.02862548828125,
-0.049896240234375,
0.01105499267578125,
0.0188140869140625,
0.02288818359375,
0.060272216796875,
-0.0160369873046875,
0.00849151611328125,
0.0576171875,
-0.036346435546875,
0.0479736328125,
0.04632568359375,
-0.0258331298828125,
-0.0328369140625,
0.047882080078125,
0.0056304931640625,
0.006015777587890625,
0.035675048828125,
0.025299072265625,
-0.017333984375,
-0.0118255615234375,
-0.0570068359375,
0.017425537109375,
-0.052764892578125,
-0.013824462890625,
-0.080078125,
-0.00409698486328125,
-0.048095703125,
-0.0164031982421875,
0.004009246826171875,
-0.045257568359375,
-0.0509033203125,
-0.0027408599853515625,
0.039825439453125,
0.041290283203125,
0.0023136138916015625,
0.036224365234375,
-0.058807373046875,
0.0204315185546875,
0.018798828125,
-0.0032482147216796875,
0.00533294677734375,
-0.04156494140625,
-0.0059814453125,
0.0085296630859375,
-0.03216552734375,
-0.08282470703125,
0.035430908203125,
0.00848388671875,
0.03167724609375,
0.035186767578125,
0.0172576904296875,
0.050689697265625,
-0.03314208984375,
0.086181640625,
-0.0029296875,
-0.055694580078125,
0.060150146484375,
-0.043060302734375,
0.031982421875,
0.04351806640625,
0.03240966796875,
-0.06512451171875,
-0.02386474609375,
-0.048187255859375,
-0.062347412109375,
0.0635986328125,
0.01421356201171875,
0.01861572265625,
-0.009185791015625,
0.03631591796875,
0.0038394927978515625,
0.01324462890625,
-0.062042236328125,
-0.0296630859375,
-0.0245513916015625,
-0.00942230224609375,
0.00970458984375,
-0.01514434814453125,
-0.01239013671875,
-0.0232391357421875,
0.041168212890625,
-0.00963592529296875,
0.037322998046875,
0.0225067138671875,
0.007602691650390625,
0.0021038055419921875,
0.0165252685546875,
0.05511474609375,
0.042633056640625,
-0.0252227783203125,
0.005832672119140625,
0.01004791259765625,
-0.030548095703125,
-0.004543304443359375,
0.01384735107421875,
-0.01535797119140625,
-0.0151519775390625,
0.035125732421875,
0.058502197265625,
0.003025054931640625,
-0.054595947265625,
0.029052734375,
-0.02374267578125,
0.006458282470703125,
-0.0290374755859375,
0.0252532958984375,
0.019622802734375,
0.0206756591796875,
0.0241241455078125,
-0.0014486312866210938,
0.0213470458984375,
-0.052215576171875,
0.0039215087890625,
0.01412200927734375,
-0.010589599609375,
-0.0243377685546875,
0.043243408203125,
0.023468017578125,
-0.04229736328125,
0.047149658203125,
-0.027374267578125,
-0.0295257568359375,
0.05877685546875,
0.058013916015625,
0.05963134765625,
-0.007049560546875,
0.0193023681640625,
0.04046630859375,
0.0219573974609375,
0.00699615478515625,
0.03607177734375,
-0.0187225341796875,
-0.046051025390625,
-0.0008997917175292969,
-0.048980712890625,
-0.0228424072265625,
0.01280975341796875,
-0.045501708984375,
0.02264404296875,
-0.04168701171875,
-0.00565338134765625,
0.003658294677734375,
0.0037021636962890625,
-0.042144775390625,
0.01373291015625,
-0.00691986083984375,
0.07916259765625,
-0.08087158203125,
0.037445068359375,
0.07000732421875,
-0.0513916015625,
-0.05963134765625,
-0.00445556640625,
0.0080718994140625,
-0.035125732421875,
0.0390625,
0.0213470458984375,
0.0101776123046875,
-0.0001710653305053711,
-0.056610107421875,
-0.062103271484375,
0.09429931640625,
0.00742340087890625,
-0.0242156982421875,
-0.0162200927734375,
-0.0166168212890625,
0.0477294921875,
-0.038726806640625,
0.058380126953125,
0.030364990234375,
0.036468505859375,
0.006256103515625,
-0.06854248046875,
0.019744873046875,
-0.039764404296875,
-0.00003337860107421875,
-0.0108489990234375,
-0.0557861328125,
0.081787109375,
-0.0238037109375,
-0.026458740234375,
0.021270751953125,
0.056121826171875,
0.01470947265625,
0.0259857177734375,
0.027099609375,
0.0333251953125,
0.06805419921875,
0.0007262229919433594,
0.0775146484375,
-0.026947021484375,
0.007228851318359375,
0.09112548828125,
-0.00152587890625,
0.0543212890625,
0.0299072265625,
-0.0167388916015625,
0.0445556640625,
0.07171630859375,
0.0035114288330078125,
0.029205322265625,
0.0037364959716796875,
-0.0078887939453125,
0.00269317626953125,
-0.020538330078125,
-0.0291290283203125,
0.0229034423828125,
0.01367950439453125,
-0.0187530517578125,
-0.0016994476318359375,
0.016082763671875,
0.0212860107421875,
-0.0018777847290039062,
-0.0008020401000976562,
0.056060791015625,
0.0034542083740234375,
-0.062744140625,
0.040802001953125,
0.00730133056640625,
0.047760009765625,
-0.047637939453125,
-0.018035888671875,
-0.02679443359375,
-0.0111083984375,
-0.01380157470703125,
-0.06494140625,
0.025665283203125,
-0.0037860870361328125,
-0.0352783203125,
-0.01904296875,
0.046783447265625,
-0.03021240234375,
-0.0269622802734375,
0.00496673583984375,
0.0296783447265625,
0.04241943359375,
0.01220703125,
-0.050048828125,
0.0108489990234375,
0.0048980712890625,
-0.0251007080078125,
0.01050567626953125,
0.037506103515625,
0.0012941360473632812,
0.055938720703125,
0.04302978515625,
0.0012769699096679688,
-0.02197265625,
-0.020263671875,
0.0709228515625,
-0.058563232421875,
-0.046356201171875,
-0.056304931640625,
0.053253173828125,
-0.005435943603515625,
-0.0433349609375,
0.05169677734375,
0.038848876953125,
0.05206298828125,
-0.0004584789276123047,
0.04803466796875,
-0.035400390625,
0.03948974609375,
-0.036895751953125,
0.048797607421875,
-0.037078857421875,
0.01873779296875,
-0.0182647705078125,
-0.05181884765625,
0.0020427703857421875,
0.05731201171875,
-0.008880615234375,
0.0027370452880859375,
0.058807373046875,
0.07159423828125,
0.004669189453125,
0.01715087890625,
0.00923919677734375,
0.0208587646484375,
0.01316070556640625,
0.04376220703125,
0.05963134765625,
-0.037841796875,
0.02862548828125,
-0.020660400390625,
-0.037994384765625,
-0.00726318359375,
-0.062408447265625,
-0.07244873046875,
-0.059051513671875,
-0.01047515869140625,
-0.03759765625,
0.00984954833984375,
0.088134765625,
0.047332763671875,
-0.0576171875,
-0.03204345703125,
0.0174560546875,
0.0012769699096679688,
-0.01438140869140625,
-0.023406982421875,
0.01456451416015625,
-0.007244110107421875,
-0.0491943359375,
0.0287017822265625,
0.0013170242309570312,
0.01995849609375,
-0.025665283203125,
0.006023406982421875,
-0.0204620361328125,
0.0105438232421875,
0.041259765625,
0.029266357421875,
-0.056243896484375,
-0.0247344970703125,
0.0243377685546875,
-0.006877899169921875,
-0.004734039306640625,
0.043670654296875,
-0.0650634765625,
0.03546142578125,
0.039306640625,
0.0238189697265625,
0.0287322998046875,
0.00531005859375,
0.0367431640625,
-0.047576904296875,
-0.004825592041015625,
0.021087646484375,
0.0266876220703125,
0.024627685546875,
-0.0570068359375,
0.0322265625,
0.021759033203125,
-0.049407958984375,
-0.06640625,
0.00484466552734375,
-0.08404541015625,
-0.039276123046875,
0.08880615234375,
-0.0164947509765625,
-0.0261993408203125,
-0.0103912353515625,
-0.0489501953125,
0.00754547119140625,
-0.051666259765625,
0.047576904296875,
0.0638427734375,
-0.036956787109375,
-0.0091400146484375,
-0.03619384765625,
0.03070068359375,
0.014007568359375,
-0.07073974609375,
0.01361846923828125,
0.057159423828125,
0.02520751953125,
0.0240936279296875,
0.06610107421875,
0.0195770263671875,
0.019317626953125,
0.006107330322265625,
-0.00010412931442260742,
-0.021240234375,
-0.036285400390625,
-0.0156707763671875,
0.01000213623046875,
-0.018524169921875,
-0.01404571533203125
]
] |
lyogavin/Anima-7B-100K | 2023-09-16T01:59:42.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama2",
"100k",
"7b",
"custom_code",
"en",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | lyogavin | null | null | lyogavin/Anima-7B-100K | 14 | 7,605 | transformers | 2023-09-14T14:47:16 | ---
license: apache-2.0
language:
- en
tags:
- llama2
- 100k
- 7b
---
Anima LLM supporting 100K input token length. It's trained based on Llama2 7B, so the license support commercial use!
We carefully curated long QA training dataset from 30k to 100k length to train this model. We also made a lot of memory optimizations to make it scale to 100k tokens.
## How to train/infer?
#### install dependencies
```bash
# Please update the path of `CUDA_HOME`
export CUDA_HOME=/usr/local/cuda-11.8
pip install transformers==4.31.0
pip install sentencepiece
pip install ninja
pip install flash-attn --no-build-isolation
pip install git+https://github.com/HazyResearch/flash-attention.git#subdirectory=csrc/rotary
pip install git+https://github.com/HazyResearch/flash-attention.git#subdirectory=csrc/xentropy
pip install evaluate
pip install git+https://github.com/huggingface/peft.git@v0.4.0
pip install wandb
```
#### inference
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
base_model = "lyogavin/Anima-7B-100K"
tokenizer = AutoTokenizer.from_pretrained(base_model)
model = AutoModelForCausalLM.from_pretrained(
base_model,
torch_dtype=torch.float16,
trust_remote_code=True,
device_map="auto",
)
model.eval()
prompt = "Where is the capital of US?"
inputs = tokenizer(prompt, return_tensors="pt")
inputs['input_ids'] = inputs['input_ids'].cuda()
inputs['attention_mask'] = inputs['attention_mask'].cuda()
# Generate
generate_ids = model.generate(**inputs, max_new_tokens=30,
only_last_logit=True, # to save memory
use_cache=False, # when run into OOM, enable this can save memory
xentropy=True)
output = tokenizer.batch_decode(generate_ids,
skip_special_tokens=True,
clean_up_tokenization_spaces=False)[0]
```
#### Training
```bash
./run_longer_training.sh
```
## Evaluations
There's almost none evaluation dataset designed for 100k tokens. So we designed/curated some dataset for this model. We compared this model and several other public/private models.
#### 1. longchat topic retrieval
| Model | Accuracy |
|-------------------|---------|
| Claude2 | 0.9 |
| together llama2 32k | 0.15 |
| longchat 32k 1.5 | 0.05 |
| Anima 100K | 0.5 |
#### 2. longchat number retrieval
| Model | Accuracy |
|-------------------|---------|
| Claude2 | 0.85 |
| together llama2 32k | 0.2 |
| longchat 32k 1.5 | 0.05 |
| Anima 100K | 0.45 |
#### 3. Narrative QA in zeroscore
| Model | F1 |
|-------------------|---------|
| Claude2 | 0.6187 |
| together llama2 32k | 0.3833 |
| longchat 32k 1.5 | 0.2416 |
| Anima 100K | 0.4919 |
## Github
Github repo is [here](https://github.com/lyogavin/Anima/tree/main/anima_100k) | 2,981 | [
[
-0.02056884765625,
-0.038970947265625,
0.035308837890625,
0.017974853515625,
-0.0165557861328125,
0.0153045654296875,
-0.00006604194641113281,
-0.0263824462890625,
0.0103302001953125,
0.0179901123046875,
-0.044647216796875,
-0.0428466796875,
-0.040924072265625,
0.0028076171875,
-0.0333251953125,
0.07061767578125,
0.00749969482421875,
0.005985260009765625,
-0.009185791015625,
-0.0299835205078125,
-0.036163330078125,
-0.0479736328125,
-0.073486328125,
-0.01007843017578125,
0.032196044921875,
0.022613525390625,
0.03546142578125,
0.044158935546875,
0.027435302734375,
0.020751953125,
-0.01593017578125,
0.00457763671875,
-0.04302978515625,
-0.01511383056640625,
0.01247406005859375,
-0.051849365234375,
-0.040496826171875,
-0.001979827880859375,
0.035125732421875,
0.007213592529296875,
0.00597381591796875,
0.040008544921875,
-0.0188446044921875,
0.037994384765625,
-0.0484619140625,
0.01509857177734375,
-0.023345947265625,
-0.0017023086547851562,
-0.0094146728515625,
0.0126190185546875,
-0.0155029296875,
-0.024749755859375,
0.0066070556640625,
-0.043731689453125,
-0.007564544677734375,
0.012847900390625,
0.09906005859375,
0.0302734375,
-0.039642333984375,
-0.020965576171875,
-0.0292205810546875,
0.058319091796875,
-0.07171630859375,
0.0131988525390625,
0.033447265625,
0.01348114013671875,
0.007427215576171875,
-0.0643310546875,
-0.040618896484375,
-0.0156707763671875,
-0.0039825439453125,
0.0093536376953125,
-0.035797119140625,
0.005519866943359375,
0.037445068359375,
0.033050537109375,
-0.033111572265625,
0.01739501953125,
-0.04388427734375,
-0.01500701904296875,
0.052276611328125,
0.0282440185546875,
0.010284423828125,
-0.02435302734375,
-0.035186767578125,
-0.00926971435546875,
-0.038330078125,
0.028839111328125,
0.042236328125,
-0.01381683349609375,
-0.0188140869140625,
0.03643798828125,
-0.02117919921875,
0.03509521484375,
0.020233154296875,
-0.0057830810546875,
0.0290985107421875,
-0.0377197265625,
-0.0182342529296875,
0.0001552104949951172,
0.0819091796875,
0.01434326171875,
0.0126495361328125,
0.004322052001953125,
-0.0177154541015625,
0.0173187255859375,
0.020416259765625,
-0.069091796875,
-0.001811981201171875,
0.035675048828125,
-0.028717041015625,
-0.0264739990234375,
-0.00850677490234375,
-0.037933349609375,
-0.0238037109375,
-0.01099395751953125,
0.04241943359375,
-0.036468505859375,
-0.01551055908203125,
0.00749969482421875,
0.004833221435546875,
0.026885986328125,
0.0139312744140625,
-0.057891845703125,
0.0225067138671875,
0.035919189453125,
0.0682373046875,
-0.0265350341796875,
-0.0279541015625,
-0.004413604736328125,
-0.0189666748046875,
-0.024200439453125,
0.043853759765625,
-0.00882720947265625,
-0.01461029052734375,
-0.0270538330078125,
0.003925323486328125,
-0.01549530029296875,
-0.0275726318359375,
0.024322509765625,
-0.0350341796875,
0.030975341796875,
-0.01485443115234375,
-0.039886474609375,
-0.02850341796875,
0.017303466796875,
-0.037139892578125,
0.09222412109375,
0.019256591796875,
-0.0643310546875,
0.0270843505859375,
-0.05828857421875,
-0.0121917724609375,
-0.0141143798828125,
-0.0064544677734375,
-0.0484619140625,
0.0005369186401367188,
0.02685546875,
0.0316162109375,
-0.021820068359375,
0.0338134765625,
-0.0178375244140625,
-0.0509033203125,
0.02484130859375,
-0.035919189453125,
0.08367919921875,
0.024993896484375,
-0.031463623046875,
0.0311126708984375,
-0.06683349609375,
-0.0143890380859375,
0.01538848876953125,
-0.01544189453125,
0.00002002716064453125,
-0.03692626953125,
-0.002285003662109375,
0.012420654296875,
0.028472900390625,
-0.044219970703125,
0.008758544921875,
-0.03900146484375,
0.0638427734375,
0.049102783203125,
0.00795745849609375,
0.02978515625,
-0.029449462890625,
0.043914794921875,
0.0169677734375,
0.03680419921875,
-0.01263427734375,
-0.034271240234375,
-0.07928466796875,
-0.0341796875,
0.041595458984375,
0.035247802734375,
-0.034454345703125,
0.04095458984375,
-0.0137939453125,
-0.037445068359375,
-0.06805419921875,
-0.0010175704956054688,
0.0307769775390625,
0.039581298828125,
0.008514404296875,
-0.0440673828125,
-0.033477783203125,
-0.071044921875,
0.01409149169921875,
-0.01454925537109375,
-0.003108978271484375,
0.0279693603515625,
0.0584716796875,
-0.020050048828125,
0.0787353515625,
-0.025299072265625,
-0.026275634765625,
-0.005359649658203125,
-0.0031795501708984375,
0.03338623046875,
0.0263214111328125,
0.05206298828125,
-0.040069580078125,
-0.0310211181640625,
-0.02545166015625,
-0.056549072265625,
0.0148162841796875,
-0.0010051727294921875,
-0.018524169921875,
0.02154541015625,
0.0124053955078125,
-0.05511474609375,
0.05487060546875,
0.0287933349609375,
-0.0404052734375,
0.03485107421875,
-0.0184478759765625,
0.0162200927734375,
-0.08966064453125,
0.0287933349609375,
-0.0145721435546875,
-0.0199737548828125,
-0.037078857421875,
-0.004665374755859375,
0.0215911865234375,
-0.0170440673828125,
-0.05059814453125,
0.06292724609375,
-0.031341552734375,
-0.004486083984375,
-0.0123748779296875,
-0.01419830322265625,
0.01131439208984375,
0.0657958984375,
-0.0106658935546875,
0.07000732421875,
0.043975830078125,
-0.0457763671875,
0.039306640625,
0.023223876953125,
-0.033050537109375,
0.0267791748046875,
-0.05804443359375,
0.01641845703125,
0.003173828125,
0.007160186767578125,
-0.09429931640625,
-0.0197296142578125,
0.01549530029296875,
-0.047088623046875,
0.012451171875,
-0.00673675537109375,
-0.042266845703125,
-0.0413818359375,
-0.0290679931640625,
0.034698486328125,
0.040618896484375,
-0.04376220703125,
0.0244293212890625,
0.0258636474609375,
0.00684356689453125,
-0.048431396484375,
-0.052703857421875,
-0.001453399658203125,
-0.033203125,
-0.044403076171875,
0.0023059844970703125,
-0.0035552978515625,
0.004138946533203125,
0.007549285888671875,
-0.0058135986328125,
-0.0061798095703125,
0.009735107421875,
0.034759521484375,
0.0281829833984375,
-0.0133819580078125,
0.0123748779296875,
-0.01280975341796875,
-0.01146697998046875,
0.00530242919921875,
0.0113525390625,
0.055267333984375,
-0.0201873779296875,
-0.0198211669921875,
-0.0599365234375,
-0.00548553466796875,
0.051605224609375,
-0.017242431640625,
0.06005859375,
0.046112060546875,
-0.0266571044921875,
0.01531219482421875,
-0.060333251953125,
-0.0125579833984375,
-0.034393310546875,
0.034149169921875,
-0.013153076171875,
-0.056396484375,
0.06658935546875,
0.01525115966796875,
0.007114410400390625,
0.04840087890625,
0.04779052734375,
0.018218994140625,
0.07489013671875,
0.036224365234375,
-0.032470703125,
0.046966552734375,
-0.04364013671875,
0.00656890869140625,
-0.05914306640625,
-0.0174102783203125,
-0.020965576171875,
-0.016265869140625,
-0.0279998779296875,
-0.017730712890625,
0.0204620361328125,
-0.00782012939453125,
-0.043182373046875,
0.0280914306640625,
-0.048248291015625,
0.0141754150390625,
0.054718017578125,
0.007015228271484375,
0.01300048828125,
-0.00499725341796875,
-0.0005044937133789062,
0.0084228515625,
-0.055908203125,
-0.0261077880859375,
0.095703125,
0.056671142578125,
0.0504150390625,
-0.01351165771484375,
0.05706787109375,
0.004688262939453125,
0.0189971923828125,
-0.06561279296875,
0.025299072265625,
0.0107574462890625,
-0.05926513671875,
-0.020599365234375,
-0.0307159423828125,
-0.07342529296875,
-0.0008268356323242188,
-0.00882720947265625,
-0.0501708984375,
0.0306549072265625,
-0.01229095458984375,
-0.04522705078125,
0.011383056640625,
-0.062255859375,
0.06793212890625,
-0.0221405029296875,
-0.0225067138671875,
-0.0031032562255859375,
-0.038177490234375,
0.037811279296875,
-0.00650787353515625,
0.0026836395263671875,
-0.0126800537109375,
-0.002651214599609375,
0.08148193359375,
-0.026885986328125,
0.0633544921875,
-0.00545501708984375,
0.008087158203125,
0.0307464599609375,
-0.020294189453125,
0.0299224853515625,
0.0206146240234375,
-0.0101776123046875,
0.027740478515625,
0.0156402587890625,
-0.03216552734375,
-0.0296478271484375,
0.04351806640625,
-0.0806884765625,
-0.03485107421875,
-0.049163818359375,
-0.040252685546875,
-0.00505828857421875,
0.01480865478515625,
0.035980224609375,
0.0202789306640625,
0.0099639892578125,
0.01708984375,
0.044952392578125,
-0.0277862548828125,
0.037078857421875,
0.0268707275390625,
-0.004062652587890625,
-0.048919677734375,
0.053131103515625,
0.01387786865234375,
0.004291534423828125,
-0.006160736083984375,
-0.0030651092529296875,
-0.036163330078125,
-0.0273590087890625,
-0.036407470703125,
0.0267181396484375,
-0.040008544921875,
-0.03643798828125,
-0.048553466796875,
-0.036956787109375,
-0.0296630859375,
0.01313018798828125,
-0.0498046875,
-0.0270843505859375,
-0.046417236328125,
-0.01198577880859375,
0.04345703125,
0.0570068359375,
0.02154541015625,
0.03466796875,
-0.032501220703125,
0.02203369140625,
0.02313232421875,
0.0263519287109375,
0.020904541015625,
-0.06365966796875,
-0.0165863037109375,
-0.0022563934326171875,
-0.027313232421875,
-0.06646728515625,
0.05206298828125,
-0.00836181640625,
0.039093017578125,
0.03436279296875,
0.004367828369140625,
0.06298828125,
-0.0051116943359375,
0.058837890625,
0.0225677490234375,
-0.06219482421875,
0.0391845703125,
-0.007793426513671875,
0.03192138671875,
0.05145263671875,
0.036346435546875,
-0.0240631103515625,
-0.025177001953125,
-0.06646728515625,
-0.07177734375,
0.063232421875,
0.0252685546875,
0.01104736328125,
0.005321502685546875,
0.0458984375,
-0.0002334117889404297,
0.0266571044921875,
-0.059173583984375,
-0.037506103515625,
-0.01015472412109375,
-0.020294189453125,
-0.00048351287841796875,
-0.0310211181640625,
-0.00923919677734375,
-0.0279541015625,
0.045257568359375,
0.004924774169921875,
0.034271240234375,
0.01343536376953125,
-0.0019779205322265625,
0.00432586669921875,
0.00708770751953125,
0.0303497314453125,
0.0701904296875,
-0.020111083984375,
-0.007472991943359375,
0.033935546875,
-0.038177490234375,
0.00865936279296875,
0.0005283355712890625,
-0.01520538330078125,
-0.01483154296875,
0.0197906494140625,
0.0875244140625,
0.010345458984375,
-0.038177490234375,
0.048553466796875,
-0.0267181396484375,
-0.0226898193359375,
-0.024871826171875,
0.01181793212890625,
0.007099151611328125,
0.0286102294921875,
0.035003662109375,
-0.0200042724609375,
0.004608154296875,
-0.045074462890625,
0.00753021240234375,
0.0199737548828125,
-0.033843994140625,
-0.02069091796875,
0.06719970703125,
-0.0020046234130859375,
-0.016448974609375,
0.054718017578125,
-0.0196533203125,
-0.03668212890625,
0.06341552734375,
0.02056884765625,
0.05865478515625,
-0.0021266937255859375,
0.00743865966796875,
0.0474853515625,
0.00910186767578125,
-0.0167999267578125,
0.0219879150390625,
0.011444091796875,
-0.032867431640625,
-0.017669677734375,
-0.05706787109375,
-0.032501220703125,
0.0222625732421875,
-0.05706787109375,
0.0277557373046875,
-0.042327880859375,
-0.03369140625,
-0.01395416259765625,
0.0465087890625,
-0.06207275390625,
0.0206146240234375,
-0.00196075439453125,
0.060272216796875,
-0.05218505859375,
0.059722900390625,
0.04217529296875,
-0.04339599609375,
-0.06683349609375,
-0.0301666259765625,
-0.01904296875,
-0.0833740234375,
0.0386962890625,
0.0177459716796875,
0.033416748046875,
0.007049560546875,
-0.040008544921875,
-0.07861328125,
0.11065673828125,
0.036590576171875,
-0.04241943359375,
-0.009368896484375,
0.0021114349365234375,
0.050994873046875,
-0.0267486572265625,
0.041229248046875,
0.0330810546875,
0.0197906494140625,
0.01500701904296875,
-0.0804443359375,
0.0169525146484375,
-0.0177154541015625,
-0.006988525390625,
-0.01351165771484375,
-0.0789794921875,
0.0728759765625,
-0.0152587890625,
-0.0200042724609375,
0.01412200927734375,
0.04742431640625,
0.03717041015625,
0.032562255859375,
0.02996826171875,
0.049346923828125,
0.056243896484375,
-0.0014352798461914062,
0.07427978515625,
-0.034332275390625,
0.03900146484375,
0.053680419921875,
0.016082763671875,
0.052490234375,
0.0200042724609375,
-0.01715087890625,
0.0291900634765625,
0.07415771484375,
-0.01849365234375,
0.03143310546875,
0.01303863525390625,
-0.0019474029541015625,
-0.006359100341796875,
0.00112152099609375,
-0.025115966796875,
0.0307769775390625,
0.016357421875,
-0.0234527587890625,
-0.0009131431579589844,
-0.003482818603515625,
0.007663726806640625,
-0.032135009765625,
-0.0200042724609375,
0.056243896484375,
0.013214111328125,
-0.036895751953125,
0.05804443359375,
-0.004486083984375,
0.06402587890625,
-0.045196533203125,
0.01617431640625,
-0.031005859375,
0.002681732177734375,
-0.0175933837890625,
-0.0328369140625,
0.0030384063720703125,
0.005908966064453125,
0.004608154296875,
0.0020236968994140625,
0.046173095703125,
-0.00685882568359375,
-0.035888671875,
0.019561767578125,
0.0248260498046875,
0.01517486572265625,
-0.0043182373046875,
-0.06427001953125,
0.00335693359375,
-0.006946563720703125,
-0.050262451171875,
0.037841796875,
0.031707763671875,
-0.00685882568359375,
0.060028076171875,
0.044219970703125,
0.00554656982421875,
0.013671875,
-0.0037403106689453125,
0.0770263671875,
-0.05999755859375,
-0.0462646484375,
-0.06256103515625,
0.02972412109375,
-0.0009617805480957031,
-0.05029296875,
0.05841064453125,
0.05462646484375,
0.044281005859375,
0.0112152099609375,
0.035552978515625,
0.002429962158203125,
0.01561737060546875,
-0.0234222412109375,
0.06365966796875,
-0.04443359375,
0.0160980224609375,
0.0089111328125,
-0.059478759765625,
-0.0072479248046875,
0.04742431640625,
-0.02783203125,
0.0050201416015625,
0.03765869140625,
0.056365966796875,
-0.005397796630859375,
-0.01837158203125,
0.0242767333984375,
0.0190887451171875,
0.0126800537109375,
0.05889892578125,
0.060791015625,
-0.058807373046875,
0.04620361328125,
-0.031707763671875,
-0.0085601806640625,
-0.00719451904296875,
-0.046905517578125,
-0.08563232421875,
-0.0270843505859375,
-0.029388427734375,
-0.030364990234375,
0.0029125213623046875,
0.047454833984375,
0.058990478515625,
-0.062744140625,
-0.0234527587890625,
0.00682830810546875,
0.00933837890625,
-0.0002791881561279297,
-0.0194244384765625,
0.045196533203125,
-0.0230560302734375,
-0.058563232421875,
0.0060882568359375,
0.00023615360260009766,
0.00315093994140625,
-0.0302276611328125,
-0.007099151611328125,
-0.0212860107421875,
0.0051422119140625,
0.0251007080078125,
0.0226287841796875,
-0.040863037109375,
-0.0184326171875,
0.009765625,
-0.0202789306640625,
0.0190887451171875,
0.01291656494140625,
-0.058563232421875,
0.0210113525390625,
0.0343017578125,
0.0350341796875,
0.06695556640625,
-0.0007152557373046875,
0.0197296142578125,
-0.05279541015625,
0.0122528076171875,
0.0160064697265625,
0.0166015625,
0.0137939453125,
-0.03857421875,
0.04974365234375,
0.0108642578125,
-0.04864501953125,
-0.0811767578125,
-0.01396942138671875,
-0.076416015625,
-0.007312774658203125,
0.097412109375,
-0.0220794677734375,
-0.0455322265625,
0.004169464111328125,
-0.0234832763671875,
0.032135009765625,
-0.041961669921875,
0.064697265625,
0.05419921875,
-0.01152801513671875,
-0.013427734375,
-0.0418701171875,
0.0330810546875,
0.0234222412109375,
-0.0570068359375,
-0.01177215576171875,
0.01012420654296875,
0.04498291015625,
0.014984130859375,
0.06365966796875,
-0.00879669189453125,
0.0302734375,
-0.0002713203430175781,
0.006298065185546875,
-0.031768798828125,
-0.005535125732421875,
-0.044525146484375,
0.004486083984375,
0.00021255016326904297,
-0.03216552734375
]
] |
jondurbin/airoboros-c34b-2.2.1 | 2023-09-28T09:39:42.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:jondurbin/airoboros-2.2.1",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | jondurbin | null | null | jondurbin/airoboros-c34b-2.2.1 | 9 | 7,603 | transformers | 2023-09-19T20:22:58 | ---
license: llama2
datasets:
- jondurbin/airoboros-2.2.1
---
### Overview
Another experimental model, using mostly sythetic data generated by [airoboros](https://github.com/jondurbin/airoboros)
This is essentially a minor "fix" branch of [airoboros-c34b-2.2](https://hf.co/jondurbin/airoboros-c34b-2.2) with a updates, primarily:
- [re-generated writing responses](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#re-generated-writing-responses)
- [longer contextual blocks](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#longer-contextual-blocks)
- [removal of "rp" data](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#rp-category-removed)
- [(less aggressive) de-censoring](https://huggingface.co/datasets/jondurbin/airoboros-2.2.1#de-censoring)
- 5 epochs instead of 3
This is a fairly general purpose model, but focuses heavily on instruction following, rather than casual chat/roleplay.
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 prompt format:
```
A chat.
USER: {prompt}
ASSISTANT:
```
The default system prompt ("A chat.") was used for most of the prompts, however it also included a wide sampling of responses with other prompts, particularly in "stylized\_response", "rp", "gtkm", etc.
Here's another example:
```
A chat between Bob (aka USER) and Tom (aka ASSISTANT). Tom is an extremely intelligent 18th century bookkeeper, who speaks loquaciously.
USER: {prompt}
ASSISTANT:
```
And chat scenario that wouldn't require USER/ASSISTANT (but should use stopping criteria to prevent the model from speaking on your behalf).
```
A chat between old friends: Timmy and Tommy.
{description of characters}
{setting for the chat}
Timmy: *takes a big sip from his coffee* "Ah, sweet, delicious, magical coffee."
Tommy:
```
__*I strongly suggest adding stopping criteria/early inference stopping on "USER:", and/or whatever names you specify in the system prompt.*__
### Fine tuning info
https://wandb.ai/jondurbin/airoboros-c34b-2.2.1/runs/07b08z7m?workspace=user-jondurbin
### 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
```
#### Summarization
500 samples have been included from [this dataset](https://huggingface.co/datasets/mattpscott/airoboros-summarization), using the same format as contextual question answering, for example:
```
BEGININPUT
{text to summarize}
ENDINPUT
BEGININSTRUCTION
Summarize the input in around 130 words.
ENDINSTRUCTION
```
#### Getting longer responses
You can use a few techniques to get longer responses.
Detailed prompts, with explicit instruction for word count:
```
Please compose a narrative set in the heart of an ancient library, steeped in the scent of old parchment and ink. The protagonist should be a young scholar who is dedicated to studying the art of storytelling and its evolution throughout history. In her pursuit of knowledge, she stumbles upon a forgotten tome that seems to possess an unusual aura. This book has the ability to bring stories to life, literally manifesting characters and scenarios from within its pages into reality.
The main character must navigate through various epochs of storytelling - from oral traditions of tribal societies, through medieval minstrels' tales, to modern-day digital narratives - as they come alive around her. Each era presents its unique challenges and lessons about the power and impact of stories on human civilization.
One such character could be a sentient quill pen, who was once used by renowned authors of yesteryears and now holds their wisdom and experiences. It becomes her mentor, guiding her through this journey with witty remarks and insightful commentary.
Ensure that your tale encapsulates the thrill of adventure, the beauty of learning, and the profound connection between humans and their stories. All characters involved should be non-human entities. Feel free to explore creative liberties but maintain the mentioned elements.
Your response should be approximately 2300 words.
```
Or, a simpler example:
```
Please create a long, detailed story about a dragon in an old growth forest who, for some reason, begins speaking the words of the source code of linux.
```
There are a few examples of next chapter completion as well, e.g.:
```
Write the next chapter of a historical fiction novel set in Paris during the 20th century.
Here's a summary of the previous chapter:
In the vibrant city of Paris, amid the tumultuous changes of the 20th century, our protagonist Margot, an aspiring fashion designer, has just secured an apprenticeship at a prestigious couture house. She meets Lucien, a charming journalist who covers the fashion industry. Together they navigate the ever-changing world of fashion and society, uncovering secrets that reveal the intricate links between style, politics, and culture. As the chapter concludes, they decide to delve deeper into the hidden corners of the fashion world to unravel its mysteries.
Requirements for the next chapter:
1. Character Development of Margot and Lucien:
- Margot's Evolution: Unfold more about Margot's past, her dreams of revolutionizing fashion, and her struggle to establish herself in a male-dominated industry. Illustrate her growing expertise, innovative ideas, and increasing dependence on Lucien.
- Lucien's Complexity: Introduce uncertainties surrounding Lucien's background and real motives. Increase suspense by suggesting undisclosed information he possesses, while also highlighting his wit and perceptiveness.
2. Exploration of Paris and the Couture House:
- Paris: Elaborate their journey through the bustling streets of Paris, including encounters with iconic figures, social unrest, and relics from different eras of French history.
- The Couture House: Expand on the grandeur of the couture house they work in, filled with artistic masterpieces, intense competition, and cryptic notes hinting at a scandalous past.
3. Emergence of the Subplot: The Lost Collection:
- Discovery: Have Margot and Lucien stumble upon a secret vault containing a lost collection designed before World War II, raising new questions about the previous owner and the influence of war on fashion.
- Revelation: Capture their shock as they realize the designs were plagiarized, the potential repercussions, and the opportunities it presents for Margot's career.
- Twist: End with a twist that suggests there are other stolen collections across Paris, setting up their new mission.
Your response should be approximately 650 words.
```
#### 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 2.2 models are built on top of llama-2/codellama.
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 mostly 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. | 21,108 | [
[
-0.028076171875,
-0.078369140625,
0.03570556640625,
0.0004131793975830078,
-0.0026912689208984375,
-0.024688720703125,
-0.008453369140625,
-0.035675048828125,
0.0232086181640625,
0.045501708984375,
-0.07000732421875,
-0.03411865234375,
-0.0290679931640625,
0.01416778564453125,
-0.0168609619140625,
0.08343505859375,
0.0029048919677734375,
-0.01629638671875,
-0.015655517578125,
-0.0013380050659179688,
-0.052642822265625,
-0.04510498046875,
-0.05126953125,
-0.0038738250732421875,
0.04168701171875,
0.032989501953125,
0.044708251953125,
0.046142578125,
0.019989013671875,
0.0239410400390625,
-0.0193023681640625,
0.025665283203125,
-0.047698974609375,
0.0271759033203125,
-0.01018524169921875,
-0.0294036865234375,
-0.03839111328125,
0.0026988983154296875,
0.032562255859375,
0.05328369140625,
-0.00023949146270751953,
0.01300048828125,
0.00701904296875,
0.03759765625,
-0.03607177734375,
0.00672149658203125,
-0.004093170166015625,
-0.0031871795654296875,
-0.006038665771484375,
-0.017486572265625,
-0.038970947265625,
-0.0172119140625,
0.01003265380859375,
-0.061370849609375,
0.0110321044921875,
0.0247650146484375,
0.065673828125,
0.004283905029296875,
-0.03363037109375,
-0.018402099609375,
-0.047088623046875,
0.061370849609375,
-0.051666259765625,
0.01517486572265625,
0.041290283203125,
0.0270538330078125,
-0.0216217041015625,
-0.06689453125,
-0.048370361328125,
-0.0148468017578125,
-0.019775390625,
0.015106201171875,
-0.0153961181640625,
-0.01177215576171875,
0.03338623046875,
0.013458251953125,
-0.054412841796875,
-0.018768310546875,
-0.044647216796875,
-0.01093292236328125,
0.040863037109375,
0.03839111328125,
0.03546142578125,
-0.032379150390625,
-0.03179931640625,
-0.0113067626953125,
-0.040283203125,
0.017974853515625,
0.02191162109375,
0.0262298583984375,
-0.0199127197265625,
0.044586181640625,
-0.0174713134765625,
0.05010986328125,
0.00408935546875,
-0.0167694091796875,
-0.0009441375732421875,
-0.03424072265625,
-0.0196533203125,
-0.0123443603515625,
0.0537109375,
0.05657958984375,
0.0197296142578125,
0.0078582763671875,
-0.0018529891967773438,
-0.01383209228515625,
0.01959228515625,
-0.060302734375,
-0.025115966796875,
0.0303802490234375,
-0.056182861328125,
-0.0171356201171875,
-0.01016998291015625,
-0.053955078125,
-0.02734375,
-0.0185089111328125,
0.027069091796875,
-0.0390625,
-0.00969696044921875,
0.01202392578125,
-0.032745361328125,
0.01004791259765625,
0.03997802734375,
-0.07940673828125,
0.028778076171875,
0.02191162109375,
0.0531005859375,
0.0016450881958007812,
-0.0170440673828125,
-0.016082763671875,
0.00965118408203125,
-0.0187530517578125,
0.05523681640625,
-0.04156494140625,
-0.039459228515625,
-0.018218994140625,
0.0284576416015625,
0.00042319297790527344,
-0.0222930908203125,
0.037261962890625,
-0.02691650390625,
0.0482177734375,
-0.040771484375,
-0.03857421875,
-0.0204010009765625,
0.0160064697265625,
-0.03594970703125,
0.0692138671875,
0.0120697021484375,
-0.06329345703125,
-0.0040283203125,
-0.06011962890625,
-0.01393890380859375,
-0.0039215087890625,
-0.0045623779296875,
0.0097198486328125,
-0.012969970703125,
0.0131988525390625,
0.034393310546875,
-0.0224761962890625,
0.01197052001953125,
-0.0283660888671875,
-0.02020263671875,
0.02154541015625,
-0.0237579345703125,
0.09173583984375,
0.0123443603515625,
-0.01190948486328125,
-0.00665283203125,
-0.04998779296875,
-0.01287841796875,
0.01532745361328125,
-0.016845703125,
-0.0191650390625,
-0.01123809814453125,
0.01200103759765625,
0.0004112720489501953,
0.0229644775390625,
-0.044189453125,
0.0323486328125,
-0.0276031494140625,
0.037628173828125,
0.0489501953125,
0.021575927734375,
0.026885986328125,
-0.05426025390625,
0.04290771484375,
-0.00946044921875,
0.0177764892578125,
-0.045196533203125,
-0.0396728515625,
-0.056610107421875,
0.0029201507568359375,
0.00936126708984375,
0.0587158203125,
-0.045928955078125,
0.025177001953125,
0.0015468597412109375,
-0.037994384765625,
-0.012451171875,
-0.01776123046875,
0.0283355712890625,
0.060791015625,
0.03985595703125,
-0.00316619873046875,
-0.043914794921875,
-0.0589599609375,
0.007293701171875,
-0.03143310546875,
-0.008514404296875,
0.03985595703125,
0.038116455078125,
-0.011260986328125,
0.07440185546875,
-0.07208251953125,
-0.0015859603881835938,
-0.0181732177734375,
-0.002185821533203125,
0.005092620849609375,
0.05889892578125,
0.039306640625,
-0.06353759765625,
-0.0139923095703125,
-0.0191497802734375,
-0.061248779296875,
-0.005962371826171875,
-0.015838623046875,
-0.02337646484375,
0.0081787109375,
0.03973388671875,
-0.0406494140625,
0.035491943359375,
0.01006317138671875,
-0.045745849609375,
0.05316162109375,
-0.00798797607421875,
0.0248260498046875,
-0.10076904296875,
0.0169677734375,
-0.0120697021484375,
0.00042748451232910156,
-0.0506591796875,
0.007160186767578125,
-0.015106201171875,
-0.00732421875,
-0.032135009765625,
0.07415771484375,
-0.033294677734375,
0.02008056640625,
-0.00160980224609375,
0.015655517578125,
0.0273284912109375,
0.05291748046875,
0.01042938232421875,
0.0401611328125,
0.0260772705078125,
-0.044189453125,
0.033111572265625,
0.040069580078125,
0.0037899017333984375,
0.048187255859375,
-0.058441162109375,
0.0093536376953125,
-0.0159759521484375,
0.0306396484375,
-0.08819580078125,
-0.026885986328125,
0.043060302734375,
-0.05377197265625,
0.0215301513671875,
0.0015850067138671875,
-0.031890869140625,
-0.0236358642578125,
-0.020538330078125,
0.0181427001953125,
0.03564453125,
-0.024688720703125,
0.05816650390625,
0.0235748291015625,
-0.019744873046875,
-0.03875732421875,
-0.056915283203125,
0.0123443603515625,
-0.020263671875,
-0.05133056640625,
0.031463623046875,
-0.046539306640625,
-0.0255279541015625,
-0.00891876220703125,
0.0020008087158203125,
-0.0180511474609375,
0.0099945068359375,
0.0198974609375,
0.03057861328125,
-0.01432037353515625,
0.01119232177734375,
0.015838623046875,
0.007526397705078125,
0.005870819091796875,
-0.01092529296875,
0.03570556640625,
-0.014434814453125,
-0.007171630859375,
-0.0306549072265625,
0.046478271484375,
0.03564453125,
-0.00023686885833740234,
0.050994873046875,
0.038970947265625,
-0.040863037109375,
0.007152557373046875,
-0.03363037109375,
-0.031890869140625,
-0.03662109375,
0.0168914794921875,
-0.0284576416015625,
-0.05450439453125,
0.04742431640625,
0.018707275390625,
0.018218994140625,
0.033203125,
0.031463623046875,
-0.023773193359375,
0.0574951171875,
0.05291748046875,
0.0212860107421875,
0.024627685546875,
-0.0240936279296875,
0.0023555755615234375,
-0.06787109375,
-0.038421630859375,
-0.036102294921875,
-0.0211944580078125,
-0.035369873046875,
-0.0120391845703125,
0.0222015380859375,
0.0254974365234375,
-0.0233306884765625,
0.0360107421875,
-0.044189453125,
0.02996826171875,
0.04345703125,
0.0166778564453125,
0.0135650634765625,
-0.0257720947265625,
-0.000028371810913085938,
-0.007350921630859375,
-0.055999755859375,
-0.05682373046875,
0.0633544921875,
0.03363037109375,
0.046600341796875,
0.022216796875,
0.0538330078125,
0.0162353515625,
0.003955841064453125,
-0.040863037109375,
0.049224853515625,
-0.0047454833984375,
-0.0523681640625,
-0.0305023193359375,
-0.01151275634765625,
-0.07122802734375,
0.00843048095703125,
-0.01418304443359375,
-0.07818603515625,
0.03594970703125,
0.0120849609375,
-0.06524658203125,
0.0036468505859375,
-0.05670166015625,
0.07659912109375,
-0.002574920654296875,
-0.0264739990234375,
0.01641845703125,
-0.077880859375,
0.0224151611328125,
0.027313232421875,
0.005840301513671875,
0.0030994415283203125,
-0.0115203857421875,
0.0657958984375,
-0.026702880859375,
0.07745361328125,
-0.004497528076171875,
0.00826263427734375,
0.0307769775390625,
0.012420654296875,
0.01525115966796875,
0.009307861328125,
0.0081787109375,
-0.006282806396484375,
0.03717041015625,
-0.0262451171875,
-0.048553466796875,
0.0484619140625,
-0.05291748046875,
-0.041168212890625,
-0.0267791748046875,
-0.05426025390625,
0.00539398193359375,
0.0173797607421875,
0.0283355712890625,
0.0650634765625,
-0.0240936279296875,
0.007389068603515625,
0.06658935546875,
-0.0325927734375,
0.04266357421875,
0.042266845703125,
-0.0264129638671875,
-0.026519775390625,
0.041351318359375,
0.007556915283203125,
0.01227569580078125,
0.034576416015625,
0.0193634033203125,
-0.0187530517578125,
-0.0091705322265625,
-0.052001953125,
0.021331787109375,
-0.0634765625,
-0.0066375732421875,
-0.08477783203125,
-0.0117034912109375,
-0.047271728515625,
-0.0197601318359375,
0.00943756103515625,
-0.040802001953125,
-0.05792236328125,
-0.0047149658203125,
0.033660888671875,
0.049774169921875,
0.003780364990234375,
0.0268402099609375,
-0.057373046875,
0.0197906494140625,
0.0212554931640625,
-0.01277923583984375,
0.00504302978515625,
-0.049530029296875,
-0.0071258544921875,
0.00402069091796875,
-0.031829833984375,
-0.0810546875,
0.03826904296875,
0.00264739990234375,
0.03173828125,
0.03338623046875,
0.025360107421875,
0.045318603515625,
-0.033447265625,
0.08905029296875,
-0.005680084228515625,
-0.051849365234375,
0.0509033203125,
-0.044952392578125,
0.03289794921875,
0.0401611328125,
0.03448486328125,
-0.060333251953125,
-0.031890869140625,
-0.055511474609375,
-0.0672607421875,
0.057647705078125,
0.0177459716796875,
0.0225982666015625,
-0.01421356201171875,
0.0301666259765625,
0.0030536651611328125,
0.01062774658203125,
-0.05889892578125,
-0.03240966796875,
-0.0191497802734375,
-0.005126953125,
0.0091705322265625,
-0.0242462158203125,
-0.014556884765625,
-0.02545166015625,
0.04754638671875,
-0.0008883476257324219,
0.03228759765625,
0.024078369140625,
0.00608062744140625,
0.004390716552734375,
0.017364501953125,
0.056243896484375,
0.05206298828125,
-0.027313232421875,
-0.0029468536376953125,
0.01322174072265625,
-0.0308685302734375,
-0.00616455078125,
0.00848388671875,
-0.0055694580078125,
-0.00832366943359375,
0.03924560546875,
0.058319091796875,
0.01084136962890625,
-0.048309326171875,
0.030487060546875,
-0.02044677734375,
0.0030002593994140625,
-0.03082275390625,
0.026885986328125,
0.015350341796875,
0.0251312255859375,
0.028228759765625,
0.0004532337188720703,
0.0184783935546875,
-0.051116943359375,
0.0091094970703125,
0.0116424560546875,
-0.0177459716796875,
-0.017608642578125,
0.051116943359375,
0.019989013671875,
-0.0372314453125,
0.0389404296875,
-0.021148681640625,
-0.033172607421875,
0.060089111328125,
0.052154541015625,
0.061859130859375,
-0.01102447509765625,
0.016265869140625,
0.0399169921875,
0.02276611328125,
0.005939483642578125,
0.026275634765625,
-0.01702880859375,
-0.044830322265625,
-0.0005621910095214844,
-0.051788330078125,
-0.024017333984375,
0.01323699951171875,
-0.0484619140625,
0.0230865478515625,
-0.0313720703125,
-0.003330230712890625,
0.0008225440979003906,
0.004444122314453125,
-0.041595458984375,
0.0184783935546875,
-0.004589080810546875,
0.081787109375,
-0.08050537109375,
0.041412353515625,
0.055999755859375,
-0.050994873046875,
-0.0599365234375,
0.00377655029296875,
0.01421356201171875,
-0.03118896484375,
0.0450439453125,
0.021881103515625,
0.00667572021484375,
0.0010309219360351562,
-0.0545654296875,
-0.058807373046875,
0.09698486328125,
0.00632476806640625,
-0.03143310546875,
-0.0217132568359375,
-0.0259246826171875,
0.047698974609375,
-0.041839599609375,
0.041168212890625,
0.037750244140625,
0.038421630859375,
0.01018524169921875,
-0.06097412109375,
0.01308441162109375,
-0.037261962890625,
0.0007009506225585938,
-0.0075531005859375,
-0.0518798828125,
0.0849609375,
-0.0239410400390625,
-0.0284271240234375,
0.0307159423828125,
0.061065673828125,
0.01320648193359375,
0.0241546630859375,
0.0259857177734375,
0.03350830078125,
0.06451416015625,
-0.0017728805541992188,
0.08135986328125,
-0.0283660888671875,
0.010772705078125,
0.09259033203125,
0.0023593902587890625,
0.04931640625,
0.0202484130859375,
-0.01418304443359375,
0.049102783203125,
0.07501220703125,
0.00457763671875,
0.031524658203125,
0.0026149749755859375,
-0.01617431640625,
-0.0015764236450195312,
-0.0212860107421875,
-0.032684326171875,
0.017974853515625,
0.010406494140625,
-0.0267791748046875,
-0.002292633056640625,
0.0191497802734375,
0.025787353515625,
0.00142669677734375,
-0.00472259521484375,
0.05426025390625,
0.006343841552734375,
-0.054840087890625,
0.041717529296875,
0.003055572509765625,
0.038116455078125,
-0.050445556640625,
-0.0212554931640625,
-0.0281524658203125,
-0.00864410400390625,
-0.006591796875,
-0.057281494140625,
0.0194549560546875,
-0.0081787109375,
-0.042877197265625,
-0.021820068359375,
0.04449462890625,
-0.0361328125,
-0.0328369140625,
0.00850677490234375,
0.03228759765625,
0.03521728515625,
0.0142364501953125,
-0.04498291015625,
0.010040283203125,
0.005634307861328125,
-0.0269317626953125,
0.008941650390625,
0.03955078125,
-0.0017414093017578125,
0.05126953125,
0.03814697265625,
-0.005680084228515625,
-0.0288238525390625,
-0.01502227783203125,
0.071044921875,
-0.05584716796875,
-0.0472412109375,
-0.05859375,
0.04998779296875,
-0.0031948089599609375,
-0.042205810546875,
0.056793212890625,
0.036224365234375,
0.05523681640625,
0.0005507469177246094,
0.04742431640625,
-0.0287322998046875,
0.035888671875,
-0.04193115234375,
0.0538330078125,
-0.041412353515625,
0.0243072509765625,
-0.0210723876953125,
-0.05255126953125,
-0.0030956268310546875,
0.045745849609375,
-0.0125579833984375,
0.00984954833984375,
0.059234619140625,
0.0723876953125,
0.0092926025390625,
0.01904296875,
0.016998291015625,
0.0223541259765625,
0.00864410400390625,
0.041595458984375,
0.057373046875,
-0.032135009765625,
0.0292510986328125,
-0.018310546875,
-0.03631591796875,
-0.0108795166015625,
-0.058990478515625,
-0.07904052734375,
-0.06982421875,
-0.008392333984375,
-0.041961669921875,
0.01367950439453125,
0.07879638671875,
0.04815673828125,
-0.0523681640625,
-0.02447509765625,
0.0222625732421875,
-0.0008254051208496094,
-0.01309967041015625,
-0.022796630859375,
0.01983642578125,
-0.0027179718017578125,
-0.056793212890625,
0.02679443359375,
0.006954193115234375,
0.023468017578125,
-0.030792236328125,
0.0024662017822265625,
-0.0200347900390625,
0.0184783935546875,
0.042938232421875,
0.0227508544921875,
-0.05902099609375,
-0.0262451171875,
0.0229339599609375,
-0.00632476806640625,
-0.0032958984375,
0.047637939453125,
-0.06475830078125,
0.035064697265625,
0.03668212890625,
0.0213165283203125,
0.03466796875,
0.00710296630859375,
0.039642333984375,
-0.05084228515625,
-0.0104522705078125,
0.0261993408203125,
0.029296875,
0.0283660888671875,
-0.05926513671875,
0.0347900390625,
0.022216796875,
-0.051544189453125,
-0.06658935546875,
0.011077880859375,
-0.08056640625,
-0.0482177734375,
0.08660888671875,
-0.0111083984375,
-0.0286102294921875,
-0.01338958740234375,
-0.045745849609375,
0.00925445556640625,
-0.04248046875,
0.043060302734375,
0.0706787109375,
-0.0340576171875,
-0.0128021240234375,
-0.040283203125,
0.033599853515625,
0.01508331298828125,
-0.06268310546875,
0.01477813720703125,
0.0611572265625,
0.023040771484375,
0.019989013671875,
0.06524658203125,
0.01531982421875,
0.019775390625,
0.001068115234375,
-0.0016756057739257812,
-0.0220489501953125,
-0.041839599609375,
-0.01274871826171875,
0.01197052001953125,
-0.024749755859375,
-0.00969696044921875
]
] |
amazon/LightGPT | 2023-06-06T07:28:56.000Z | [
"transformers",
"pytorch",
"gptj",
"text-generation",
"en",
"license:apache-2.0",
"has_space",
"region:us"
] | text-generation | amazon | null | null | amazon/LightGPT | 67 | 7,600 | transformers | 2023-05-24T00:57:41 | ---
license: apache-2.0
language:
- en
pipeline_tag: text-generation
inference: false
---
# LightGPT-instruct-6B Model
LightGPT-instruct is a language model based on [GPT-J 6B](https://huggingface.co/EleutherAI/gpt-j-6b). It was instruction fine-tuned on the high quality, Apache-2.0 licensed
[OIG-small-chip2](https://github.com/LAION-AI/Open-Instruction-Generalist) instruction dataset with ~200K [training examples](https://huggingface.co/datasets/0-hero/OIG-small-chip2).
## Model Details
- **Developed by:** [AWS Contributors](https://github.com/orgs/aws-samples/teams/aws-prototype-ml-apac)
- **Model type:** Transformer-based Language Model
- **Language:** English
- **Finetuned from:** [EleutherAI / gpt-j-6b](https://huggingface.co/EleutherAI/gpt-j-6b)
- **License:** Apache 2.0
- **Contact:** [Github issues](https://huggingface.co/amazon/LightGPT/discussions/new)
## Prompting
The instruction template (adapted from the [Alpaca README.md](https://github.com/tatsu-lab/stanford_alpaca#data-release)) is used to format the prompt:
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
{instruction}
### Response:
```
Input prompt example:
```
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
How can I tell if a pomegranate is ripe?
### Response:
```
The input ends with `### Response:\n` to signal that the model should
start generating the reply.
Example of generated continuation:
```
To determine whether or not a pomegranate is ripe, you should look for signs such as color and size.
A ripe pomegranate will have deep red-orange skin with small seeds inside. You may also notice that
it feels slightly softer than other fruits when squeezed. If you're unsure about ripeness, you can
cut into one of the fruit to check its texture. The flesh should be soft but still yield slightly to
gentle pressure. Finally, smell the fruit - if it has a sweet aroma, then it's likely ready to eat!
Enjoy your delicious pomegranates!
```
## Deployment
Please refer to [Deploy models to Amazon SageMaker](https://huggingface.co/docs/sagemaker/inference).
The **example** code below also shows how to deploy LightGPT-instruct to Amazon SageMaker
```python
# pip install sagemaker==2.159.0
from sagemaker.djl_inference.model import DJLModel
#An AWS IAM role that has the permission to create SageMaker endpoint
sm_role = "arn:aws:iam::XXX"
djl_model = DJLModel(
"amazon/LightGPT",
sm_role,
dtype="fp16",
task="text-generation",
# number of gpus to partition the model across, default to use 1 GPU
number_of_partitions=1
)
# this will take a few minutes to deploy
predictor = djl_model.deploy("ml.g5.2xlarge",
initial_instance_count=1)
input_str = """Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
How can I tell if a pomegranate is ripe?
### Response:"""
data = {"inputs": input_str,
"parameters":
{
"max_new_tokens":400,
"do_sample": True,
"temperature": 0.7,
"repetition_penalty": 1.1,
"top_p": 0.8,
"top_k": 50,
"min_length": 200,
}
}
result = predictor.predict(data)
print(result[0]["generated_text"])
"""
Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
How can I tell if a pomegranate is ripe?
### Response:
Ripe pomegranates are usually easy to spot, as they will be slightly soft and give when squeezed gently.
You may also notice that the skin of the fruit has begun to turn from green to yellow-green in color.
Additionally, you should smell the aroma coming from inside the fruit; it should have a sweet fruity scent.
Lastly, check for any blemishes or bruises on the outside of the fruit. If all these signs are present,
then your pomegranate is likely ready to be picked! Enjoy your fresh produce!
**Note:** To avoid bruising, make sure to cut the stem off before picking. Otherwise, you could end up
with a bruised and unappealing piece of fruit. **Warning:** Be careful when handling and cutting
pomegranates, as they can easily bruise or break.
"""
```
## Evaluation result
| | LAMBADA PPL | LAMBADA Acc | Winogrande | Hellaswag | PIQA |
|-------------------|-------------|-------------|------------|-----------|-------|
| GPT-J | 3.99 | 69.7% | 65.3% | 66.1% | 76.5% |
| LightGPT-instruct | 4.33 | 65.0% | 64.6% | 63.9% | 75.5% |
## Limitations
See limitations of GPT-J base model [here](https://huggingface.co/EleutherAI/gpt-j-6b#limitations-and-biases).
The model may fail to follow instructions with long inputs (e.g. summarize a long text).
The model often gives incorrect answers to math and reasoning questions.
Beware of hallucinations: Outputs are often factually wrong or misleading.
Replies might look convincing (at first glance) while containing completely
made up false statements.
This model is usable only for English conversations. | 5,262 | [
[
-0.0223541259765625,
-0.07696533203125,
0.041412353515625,
0.007076263427734375,
-0.0191650390625,
-0.026885986328125,
0.0006651878356933594,
-0.03973388671875,
-0.0024623870849609375,
0.038787841796875,
-0.046356201171875,
-0.0280303955078125,
-0.035797119140625,
-0.02386474609375,
0.0002868175506591797,
0.11065673828125,
-0.0127716064453125,
0.01461029052734375,
0.01201629638671875,
0.00391387939453125,
-0.041015625,
-0.025390625,
-0.04937744140625,
-0.0197906494140625,
0.0140838623046875,
0.01305389404296875,
0.06878662109375,
0.0187225341796875,
0.036346435546875,
0.0242919921875,
-0.0221405029296875,
-0.0225677490234375,
-0.00778961181640625,
-0.024383544921875,
-0.006771087646484375,
-0.0225982666015625,
-0.044952392578125,
0.0181732177734375,
0.0150909423828125,
0.052337646484375,
-0.014892578125,
0.0251007080078125,
0.001964569091796875,
0.019500732421875,
-0.0345458984375,
0.0300140380859375,
-0.02935791015625,
0.01873779296875,
-0.00372314453125,
0.01458740234375,
-0.01416015625,
-0.027252197265625,
-0.0028553009033203125,
-0.04864501953125,
0.0191192626953125,
0.004199981689453125,
0.090087890625,
0.023529052734375,
-0.0192108154296875,
-0.007488250732421875,
-0.0592041015625,
0.061492919921875,
-0.07989501953125,
-0.00951385498046875,
0.036834716796875,
0.055908203125,
-0.01149749755859375,
-0.05303955078125,
-0.04620361328125,
-0.0294342041015625,
-0.02447509765625,
0.0191650390625,
0.0075225830078125,
-0.003826141357421875,
0.037139892578125,
0.051483154296875,
-0.048065185546875,
-0.037506103515625,
-0.020050048828125,
-0.0020580291748046875,
0.033355712890625,
0.012298583984375,
0.016326904296875,
0.01029205322265625,
-0.043548583984375,
0.00485992431640625,
-0.023529052734375,
0.00766754150390625,
0.01568603515625,
0.0140228271484375,
-0.0219268798828125,
0.046417236328125,
-0.014556884765625,
0.042022705078125,
0.004993438720703125,
-0.01262664794921875,
0.036285400390625,
-0.0384521484375,
-0.051055908203125,
-0.0205535888671875,
0.0643310546875,
0.00977325439453125,
-0.004421234130859375,
-0.0031108856201171875,
-0.0285186767578125,
0.0063323974609375,
-0.0024433135986328125,
-0.07305908203125,
-0.036712646484375,
0.025665283203125,
-0.032257080078125,
-0.00685882568359375,
0.0177001953125,
-0.055999755859375,
-0.0168914794921875,
0.01239013671875,
0.051544189453125,
-0.046661376953125,
-0.007038116455078125,
0.02569580078125,
-0.005214691162109375,
0.033782958984375,
0.0247802734375,
-0.07427978515625,
0.022247314453125,
0.030670166015625,
0.043304443359375,
0.043304443359375,
-0.0262908935546875,
-0.017852783203125,
-0.01128387451171875,
-0.0039215087890625,
0.02203369140625,
-0.0296630859375,
-0.00514984130859375,
-0.0146484375,
0.04388427734375,
-0.019561767578125,
-0.0298004150390625,
0.073974609375,
-0.027191162109375,
0.036346435546875,
-0.0081939697265625,
-0.042083740234375,
-0.041168212890625,
0.000011920928955078125,
-0.0254364013671875,
0.059783935546875,
0.030670166015625,
-0.057373046875,
0.030029296875,
-0.048919677734375,
-0.029296875,
0.006683349609375,
-0.01776123046875,
-0.058349609375,
-0.00206756591796875,
0.01203155517578125,
0.0419921875,
-0.0280303955078125,
0.0097503662109375,
0.0016717910766601562,
-0.0108795166015625,
0.01393890380859375,
-0.01099395751953125,
0.072021484375,
0.0380859375,
-0.04718017578125,
0.0099639892578125,
-0.062744140625,
0.00760650634765625,
0.0587158203125,
-0.0288848876953125,
0.01233673095703125,
0.0005426406860351562,
-0.0025539398193359375,
-0.0087738037109375,
0.0231781005859375,
-0.036773681640625,
0.028045654296875,
-0.0135498046875,
0.05035400390625,
0.02423095703125,
0.01375579833984375,
0.038116455078125,
-0.029541015625,
0.0438232421875,
0.00365447998046875,
0.049346923828125,
-0.031005859375,
-0.07232666015625,
-0.060516357421875,
0.0010080337524414062,
0.0011663436889648438,
0.07159423828125,
-0.05462646484375,
0.05267333984375,
0.0144805908203125,
-0.057037353515625,
-0.0233154296875,
-0.00818634033203125,
0.03887939453125,
0.050262451171875,
0.028533935546875,
-0.017578125,
-0.012481689453125,
-0.06781005859375,
-0.00658416748046875,
-0.0007534027099609375,
-0.00531005859375,
0.0238800048828125,
0.0592041015625,
-0.00962066650390625,
0.045989990234375,
-0.052520751953125,
-0.01200103759765625,
-0.01800537109375,
0.00969696044921875,
0.057647705078125,
0.056549072265625,
0.06524658203125,
-0.0628662109375,
-0.0233917236328125,
0.00646209716796875,
-0.058074951171875,
0.0079193115234375,
0.00576019287109375,
-0.0011739730834960938,
-0.00045228004455566406,
-0.007518768310546875,
-0.047760009765625,
0.042022705078125,
0.034423828125,
-0.06695556640625,
0.04351806640625,
-0.023590087890625,
0.0295867919921875,
-0.0645751953125,
0.01094818115234375,
-0.005908966064453125,
-0.01136016845703125,
-0.044525146484375,
0.00991058349609375,
0.00684356689453125,
-0.01079559326171875,
-0.05084228515625,
0.028167724609375,
-0.03399658203125,
0.001544952392578125,
-0.005054473876953125,
-0.03326416015625,
0.004573822021484375,
0.041900634765625,
-0.0268096923828125,
0.060882568359375,
0.037353515625,
-0.047119140625,
0.0285491943359375,
0.0166015625,
-0.0511474609375,
0.0179290771484375,
-0.05401611328125,
-0.00264739990234375,
0.0011606216430664062,
0.01483154296875,
-0.0684814453125,
-0.0216522216796875,
0.048492431640625,
-0.0333251953125,
0.006008148193359375,
-0.0281219482421875,
-0.0182952880859375,
-0.0279083251953125,
0.003353118896484375,
0.0288238525390625,
0.0645751953125,
-0.02154541015625,
0.047271728515625,
0.0182342529296875,
-0.01261138916015625,
-0.02740478515625,
-0.07012939453125,
-0.023712158203125,
-0.016937255859375,
-0.055755615234375,
0.0241241455078125,
-0.004390716552734375,
-0.006969451904296875,
0.0028705596923828125,
0.019500732421875,
-0.00812530517578125,
0.01120758056640625,
0.01322174072265625,
0.0199432373046875,
-0.0087432861328125,
-0.01873779296875,
-0.016632080078125,
-0.0172119140625,
-0.005229949951171875,
-0.0283203125,
0.047393798828125,
-0.01386260986328125,
-0.03179931640625,
-0.0228424072265625,
0.007488250732421875,
0.0247802734375,
-0.02740478515625,
0.063720703125,
0.055511474609375,
-0.03350830078125,
0.0152587890625,
-0.032684326171875,
-0.027801513671875,
-0.0288238525390625,
0.01279449462890625,
-0.023284912109375,
-0.0347900390625,
0.0440673828125,
0.0168609619140625,
0.01300048828125,
0.050689697265625,
0.032196044921875,
0.0115203857421875,
0.087890625,
0.042236328125,
-0.00318145751953125,
0.044036865234375,
-0.030242919921875,
0.001956939697265625,
-0.05419921875,
-0.0225982666015625,
-0.041259765625,
-0.0089263916015625,
-0.0106658935546875,
-0.04168701171875,
0.007236480712890625,
0.027313232421875,
-0.049407958984375,
0.021728515625,
-0.05609130859375,
0.038787841796875,
0.031524658203125,
-0.002208709716796875,
-0.0019702911376953125,
-0.0102386474609375,
-0.0164794921875,
-0.0017518997192382812,
-0.056182861328125,
-0.040557861328125,
0.054046630859375,
0.0286865234375,
0.051055908203125,
0.0005979537963867188,
0.01271820068359375,
-0.0026454925537109375,
0.0247955322265625,
-0.04095458984375,
0.0462646484375,
-0.003459930419921875,
-0.051055908203125,
-0.0037288665771484375,
-0.042572021484375,
-0.08074951171875,
0.0287322998046875,
-0.0203094482421875,
-0.06640625,
-0.000021696090698242188,
0.005947113037109375,
-0.034393310546875,
0.0307464599609375,
-0.0791015625,
0.09130859375,
-0.050537109375,
-0.02838134765625,
0.0265350341796875,
-0.04986572265625,
0.037628173828125,
0.031280517578125,
0.03045654296875,
-0.0159454345703125,
0.001983642578125,
0.05859375,
-0.062164306640625,
0.05474853515625,
-0.046356201171875,
0.0020580291748046875,
0.036712646484375,
-0.005767822265625,
0.05908203125,
-0.0011606216430664062,
0.00044727325439453125,
-0.00501251220703125,
-0.01312255859375,
-0.046417236328125,
-0.0216827392578125,
0.036773681640625,
-0.051177978515625,
-0.04144287109375,
-0.04461669921875,
-0.0136260986328125,
0.005382537841796875,
0.006305694580078125,
0.024139404296875,
-0.0076446533203125,
-0.0065460205078125,
-0.01033782958984375,
0.03265380859375,
-0.03399658203125,
0.032928466796875,
0.02581787109375,
-0.052581787109375,
-0.037994384765625,
0.0592041015625,
0.01226806640625,
0.0247344970703125,
0.007671356201171875,
0.0233917236328125,
-0.004520416259765625,
-0.050384521484375,
-0.060516357421875,
0.012969970703125,
-0.04925537109375,
-0.00237274169921875,
-0.034515380859375,
-0.01117706298828125,
-0.02197265625,
-0.0031909942626953125,
-0.01739501953125,
-0.0301055908203125,
-0.0089111328125,
-0.00322723388671875,
0.041656494140625,
0.045989990234375,
0.0433349609375,
0.0284423828125,
-0.050689697265625,
-0.00563812255859375,
0.020355224609375,
0.014495849609375,
-0.01006317138671875,
-0.059417724609375,
-0.01434326171875,
0.0011034011840820312,
-0.0260009765625,
-0.07550048828125,
0.044647216796875,
0.002094268798828125,
0.03643798828125,
0.041229248046875,
-0.022247314453125,
0.0291900634765625,
-0.031524658203125,
0.08245849609375,
0.0119781494140625,
-0.061370849609375,
0.0611572265625,
-0.037750244140625,
0.032012939453125,
0.027252197265625,
0.0265960693359375,
-0.0207977294921875,
-0.016998291015625,
-0.07159423828125,
-0.07635498046875,
0.04498291015625,
0.031890869140625,
-0.004871368408203125,
0.01413726806640625,
0.0325927734375,
0.00872039794921875,
0.02569580078125,
-0.0640869140625,
-0.035858154296875,
-0.023529052734375,
0.014984130859375,
-0.018218994140625,
-0.01461029052734375,
-0.023590087890625,
-0.0205535888671875,
0.058929443359375,
0.0017328262329101562,
0.0257720947265625,
0.0113983154296875,
0.0099334716796875,
-0.0033817291259765625,
0.028045654296875,
0.037139892578125,
0.05084228515625,
-0.033721923828125,
0.0008563995361328125,
0.0201263427734375,
-0.03375244140625,
0.01050567626953125,
0.030731201171875,
-0.0362548828125,
-0.00909423828125,
0.01019287109375,
0.0654296875,
0.0014638900756835938,
-0.032928466796875,
0.034820556640625,
-0.01091766357421875,
-0.021209716796875,
-0.031646728515625,
0.0190582275390625,
0.01145172119140625,
0.01447296142578125,
0.03173828125,
0.0110321044921875,
-0.005176544189453125,
-0.04644775390625,
-0.0160980224609375,
0.0309295654296875,
-0.0201416015625,
-0.0270843505859375,
0.0833740234375,
0.01178741455078125,
-0.0202789306640625,
0.0230560302734375,
-0.0479736328125,
-0.04449462890625,
0.07501220703125,
0.040374755859375,
0.058380126953125,
-0.021697998046875,
0.005596160888671875,
0.03216552734375,
0.037109375,
-0.00830841064453125,
0.0430908203125,
0.0233154296875,
-0.04815673828125,
-0.0157928466796875,
-0.056243896484375,
-0.00904083251953125,
0.033355712890625,
-0.015869140625,
0.0006475448608398438,
-0.05877685546875,
-0.0166473388671875,
0.00830841064453125,
0.006114959716796875,
-0.056243896484375,
0.04876708984375,
-0.0128173828125,
0.046478271484375,
-0.074951171875,
0.058380126953125,
0.06915283203125,
-0.06488037109375,
-0.06634521484375,
-0.02325439453125,
-0.0025577545166015625,
-0.04876708984375,
0.004547119140625,
0.00939178466796875,
0.03265380859375,
-0.0163726806640625,
-0.047637939453125,
-0.05010986328125,
0.10614013671875,
0.03570556640625,
-0.034149169921875,
0.027099609375,
-0.007648468017578125,
0.0252685546875,
-0.0198974609375,
0.043487548828125,
0.037200927734375,
0.05078125,
0.01806640625,
-0.05804443359375,
0.0230865478515625,
-0.0174713134765625,
-0.01074981689453125,
0.025177001953125,
-0.059234619140625,
0.11181640625,
-0.01021575927734375,
-0.021026611328125,
0.0261077880859375,
0.03643798828125,
-0.0038433074951171875,
0.0165557861328125,
0.0290985107421875,
0.0584716796875,
0.06134033203125,
-0.0276641845703125,
0.1044921875,
-0.0140380859375,
0.037200927734375,
0.06439208984375,
-0.01352691650390625,
0.041778564453125,
0.0182037353515625,
-0.04156494140625,
0.005435943603515625,
0.060638427734375,
-0.025054931640625,
0.02325439453125,
0.004398345947265625,
-0.028656005859375,
0.008575439453125,
0.003673553466796875,
-0.05633544921875,
-0.004138946533203125,
0.018890380859375,
-0.053619384765625,
0.005077362060546875,
0.00266265869140625,
0.030517578125,
-0.01031494140625,
-0.0300750732421875,
0.042236328125,
-0.0021495819091796875,
-0.055023193359375,
0.0457763671875,
0.01107025146484375,
0.053680419921875,
-0.048553466796875,
-0.0015840530395507812,
-0.019866943359375,
0.033355712890625,
-0.03656005859375,
-0.0518798828125,
0.01239013671875,
-0.01485443115234375,
-0.022491455078125,
-0.0004131793975830078,
0.06256103515625,
-0.0221710205078125,
-0.0504150390625,
0.0199432373046875,
0.038909912109375,
0.034393310546875,
-0.01213836669921875,
-0.07354736328125,
-0.009765625,
0.0084381103515625,
-0.03546142578125,
0.037200927734375,
0.03399658203125,
0.005397796630859375,
0.0509033203125,
0.037506103515625,
-0.00380706787109375,
0.004871368408203125,
0.00424957275390625,
0.053680419921875,
-0.0523681640625,
-0.012420654296875,
-0.0509033203125,
0.043060302734375,
0.0030345916748046875,
-0.032501220703125,
0.06756591796875,
0.050506591796875,
0.06744384765625,
-0.032958984375,
0.0701904296875,
-0.02935791015625,
0.0260009765625,
-0.052642822265625,
0.05169677734375,
-0.0300140380859375,
0.00931549072265625,
-0.011260986328125,
-0.076171875,
-0.00376129150390625,
0.06390380859375,
-0.059326171875,
0.048553466796875,
0.051910400390625,
0.057373046875,
-0.0204620361328125,
0.006107330322265625,
0.0207672119140625,
0.025299072265625,
0.01904296875,
0.05474853515625,
0.054962158203125,
-0.044891357421875,
0.037689208984375,
-0.02960205078125,
-0.045379638671875,
-0.019561767578125,
-0.057525634765625,
-0.04296875,
-0.0271759033203125,
-0.019866943359375,
-0.0196990966796875,
0.006107330322265625,
0.046417236328125,
0.052490234375,
-0.05877685546875,
-0.002567291259765625,
-0.0194854736328125,
-0.0101470947265625,
-0.03326416015625,
-0.0180206298828125,
0.01221466064453125,
-0.0196990966796875,
-0.05889892578125,
0.01727294921875,
0.005229949951171875,
0.030792236328125,
-0.0034770965576171875,
-0.01030731201171875,
-0.0232086181640625,
-0.0096435546875,
0.0172119140625,
0.034088134765625,
-0.0276641845703125,
-0.0208740234375,
-0.033477783203125,
-0.00896453857421875,
0.008026123046875,
0.039794921875,
-0.045501708984375,
0.0091094970703125,
0.0310211181640625,
0.036529541015625,
0.04437255859375,
-0.00545501708984375,
0.030731201171875,
-0.036346435546875,
0.035736083984375,
0.032623291015625,
0.060760498046875,
0.0249481201171875,
-0.0216064453125,
0.05419921875,
0.01340484619140625,
-0.07562255859375,
-0.0176544189453125,
0.0265655517578125,
-0.06201171875,
-0.03106689453125,
0.08026123046875,
-0.0310211181640625,
-0.0438232421875,
0.004428863525390625,
-0.043853759765625,
0.0164031982421875,
-0.0294342041015625,
0.033416748046875,
0.03546142578125,
-0.007076263427734375,
-0.01323699951171875,
-0.0723876953125,
0.038238525390625,
0.059326171875,
-0.0703125,
-0.017730712890625,
0.0270538330078125,
0.0157318115234375,
0.01910400390625,
0.03765869140625,
-0.003948211669921875,
0.01439666748046875,
0.01328277587890625,
0.019622802734375,
-0.0036029815673828125,
0.00774383544921875,
-0.01493072509765625,
0.01166534423828125,
0.00820159912109375,
-0.03863525390625
]
] |
wtang06/mpt-125m-c4 | 2023-10-16T22:10:45.000Z | [
"transformers",
"pytorch",
"mpt",
"text-generation",
"custom_code",
"en",
"dataset:c4",
"license:apache-2.0",
"text-generation-inference",
"region:us"
] | text-generation | wtang06 | null | null | wtang06/mpt-125m-c4 | 1 | 7,594 | transformers | 2023-09-27T21:24:01 | ---
license: apache-2.0
language:
- en
datasets:
- c4
---
# mpt-125m-c4
## Model Description
Pretrained model for MPT-125M trained on C4 dataset
## Training data
Trained on HuggingFace C4 dataset
## Training procedure
This model was trained on C4 for ~2.5B tokens. Training time was ~1 hour with 104 A100-40gb GPUs.
## Intended Use and Limitations
This model is primarily for generating texts from a prompt. The purpose is to explore pretraining models for research. | 475 | [
[
-0.033172607421875,
-0.0335693359375,
0.048370361328125,
0.031219482421875,
-0.043060302734375,
-0.006191253662109375,
-0.0017681121826171875,
-0.0107574462890625,
0.01222991943359375,
0.01132965087890625,
-0.066162109375,
-0.0118560791015625,
-0.041046142578125,
0.00832366943359375,
-0.003520965576171875,
0.09100341796875,
-0.0197296142578125,
0.03173828125,
-0.0068359375,
-0.0011777877807617188,
-0.0255126953125,
-0.06890869140625,
-0.0494384765625,
-0.0281219482421875,
0.055267333984375,
0.0330810546875,
0.0132904052734375,
0.041717529296875,
0.04046630859375,
0.0188751220703125,
-0.00597381591796875,
-0.0328369140625,
-0.05804443359375,
-0.047882080078125,
0.0009613037109375,
-0.0259857177734375,
-0.03411865234375,
0.0250091552734375,
0.04290771484375,
0.035064697265625,
-0.0022029876708984375,
0.03924560546875,
0.007381439208984375,
0.01030731201171875,
-0.0135498046875,
0.0278472900390625,
-0.02691650390625,
0.0234527587890625,
-0.031524658203125,
-0.01276397705078125,
-0.0235443115234375,
-0.015655517578125,
0.0230712890625,
-0.0648193359375,
0.00722503662109375,
0.0162200927734375,
0.085205078125,
-0.01322174072265625,
-0.0311737060546875,
0.006023406982421875,
-0.0178680419921875,
0.058380126953125,
-0.03228759765625,
0.005096435546875,
0.03228759765625,
0.014190673828125,
0.01338958740234375,
-0.0860595703125,
-0.034759521484375,
-0.01541900634765625,
0.00673675537109375,
-0.0135650634765625,
0.01093292236328125,
0.020111083984375,
0.060546875,
0.01261138916015625,
-0.02752685546875,
-0.01033782958984375,
-0.0278472900390625,
-0.0244598388671875,
0.01517486572265625,
0.0294952392578125,
0.01953125,
-0.0184478759765625,
-0.0692138671875,
-0.000995635986328125,
-0.039398193359375,
0.0026988983154296875,
0.03619384765625,
0.0223388671875,
-0.0255126953125,
0.0535888671875,
-0.0020122528076171875,
0.025177001953125,
0.01300811767578125,
-0.0169830322265625,
0.0238037109375,
-0.041412353515625,
-0.03558349609375,
0.0013370513916015625,
0.064208984375,
0.006458282470703125,
0.0157318115234375,
0.0074615478515625,
-0.032470703125,
0.00928497314453125,
0.01531219482421875,
-0.1124267578125,
-0.0306243896484375,
-0.02532958984375,
-0.046844482421875,
-0.0298614501953125,
0.002223968505859375,
-0.0294036865234375,
0.01428985595703125,
-0.042694091796875,
0.04150390625,
-0.051544189453125,
-0.0214691162109375,
-0.009552001953125,
-0.004329681396484375,
0.0163421630859375,
0.0010852813720703125,
-0.06976318359375,
0.03363037109375,
0.031707763671875,
0.0657958984375,
-0.017730712890625,
-0.03387451171875,
-0.017791748046875,
0.0056915283203125,
-0.01511383056640625,
0.04205322265625,
-0.0291748046875,
-0.02886962890625,
-0.040130615234375,
0.029571533203125,
-0.036468505859375,
-0.00007325410842895508,
0.051727294921875,
-0.0245208740234375,
0.0396728515625,
-0.049285888671875,
-0.05743408203125,
-0.0121612548828125,
0.019744873046875,
-0.023895263671875,
0.08111572265625,
0.046112060546875,
-0.0882568359375,
0.03955078125,
-0.055572509765625,
-0.00640106201171875,
0.037628173828125,
-0.0165557861328125,
-0.0335693359375,
-0.00626373291015625,
0.00396728515625,
0.03399658203125,
-0.0025997161865234375,
0.0165557861328125,
-0.0118255615234375,
-0.04327392578125,
0.0037479400634765625,
-0.014373779296875,
0.055450439453125,
0.021728515625,
-0.02874755859375,
0.01042938232421875,
-0.0294036865234375,
-0.00390625,
0.017547607421875,
-0.0428466796875,
0.0147247314453125,
-0.041656494140625,
0.0207977294921875,
0.01421356201171875,
0.03485107421875,
-0.0697021484375,
0.01262664794921875,
-0.015655517578125,
0.0224456787109375,
0.056243896484375,
0.0021038055419921875,
0.01531982421875,
-0.033203125,
0.07037353515625,
0.005367279052734375,
0.003665924072265625,
-0.0245513916015625,
-0.0268402099609375,
-0.039215087890625,
-0.0216827392578125,
0.033447265625,
0.036651611328125,
-0.06719970703125,
0.002460479736328125,
-0.004734039306640625,
-0.030059814453125,
-0.007587432861328125,
-0.007167816162109375,
0.046630859375,
0.04193115234375,
0.017059326171875,
-0.0230255126953125,
-0.0325927734375,
-0.06243896484375,
0.0079803466796875,
-0.007495880126953125,
-0.00024771690368652344,
0.03997802734375,
0.04302978515625,
-0.0304107666015625,
0.04058837890625,
-0.03839111328125,
0.004108428955078125,
-0.01611328125,
0.0267181396484375,
-0.0014219284057617188,
0.0322265625,
0.06927490234375,
-0.032562255859375,
-0.0306396484375,
-0.0321044921875,
-0.040252685546875,
-0.0034999847412109375,
0.00177764892578125,
-0.0191192626953125,
0.00696563720703125,
0.03692626953125,
-0.056121826171875,
0.028228759765625,
0.032684326171875,
-0.037017822265625,
0.04425048828125,
-0.00769805908203125,
0.0297698974609375,
-0.09149169921875,
0.0202789306640625,
-0.010650634765625,
-0.032440185546875,
-0.01922607421875,
-0.020782470703125,
0.0067291259765625,
-0.051727294921875,
-0.043060302734375,
0.031982421875,
-0.052581787109375,
-0.00901031494140625,
-0.026031494140625,
-0.007465362548828125,
-0.01824951171875,
0.0615234375,
0.005828857421875,
0.05804443359375,
0.038299560546875,
-0.03460693359375,
0.0198516845703125,
0.034515380859375,
-0.01617431640625,
0.033477783203125,
-0.059661865234375,
0.0227508544921875,
0.007747650146484375,
0.0140838623046875,
-0.06396484375,
-0.0386962890625,
0.014984130859375,
-0.03399658203125,
0.0112152099609375,
-0.0179595947265625,
-0.0284576416015625,
-0.038238525390625,
-0.0021991729736328125,
0.072021484375,
0.027679443359375,
-0.034637451171875,
0.0299530029296875,
0.013763427734375,
-0.031585693359375,
-0.0255584716796875,
-0.039703369140625,
-0.01105499267578125,
-0.023284912109375,
-0.039764404296875,
0.03155517578125,
0.00145721435546875,
0.0272216796875,
-0.0231781005859375,
0.0267333984375,
-0.006961822509765625,
-0.0159759521484375,
0.03802490234375,
0.017913818359375,
-0.006542205810546875,
0.0011882781982421875,
-0.0157623291015625,
-0.0209808349609375,
0.0145416259765625,
0.0132904052734375,
0.0567626953125,
-0.0031223297119140625,
-0.0175628662109375,
-0.06866455078125,
0.01134490966796875,
0.0115966796875,
0.00098419189453125,
0.0611572265625,
0.0499267578125,
-0.0268707275390625,
0.0157012939453125,
-0.007598876953125,
-0.0202789306640625,
-0.035064697265625,
0.017669677734375,
-0.052642822265625,
-0.03887939453125,
0.0241851806640625,
-0.0006275177001953125,
-0.01467132568359375,
0.034576416015625,
0.048614501953125,
0.0218505859375,
0.07489013671875,
0.05511474609375,
0.0036220550537109375,
0.06195068359375,
-0.0008859634399414062,
-0.01165771484375,
-0.032745361328125,
-0.038238525390625,
-0.0172119140625,
-0.01605224609375,
-0.0244598388671875,
-0.0262603759765625,
0.0006685256958007812,
-0.0028896331787109375,
-0.048187255859375,
0.0224456787109375,
-0.036834716796875,
0.0270538330078125,
0.042510986328125,
0.0257720947265625,
-0.038177490234375,
-0.0144805908203125,
-0.0285491943359375,
-0.01126861572265625,
-0.045074462890625,
-0.0249176025390625,
0.07891845703125,
0.0401611328125,
0.04949951171875,
-0.017486572265625,
0.028106689453125,
-0.0127105712890625,
0.02655029296875,
-0.036895751953125,
0.03717041015625,
0.018524169921875,
-0.056182861328125,
-0.004306793212890625,
-0.042510986328125,
-0.05462646484375,
-0.02569580078125,
-0.018585205078125,
-0.0169525146484375,
-0.0230712890625,
0.012847900390625,
-0.00827789306640625,
0.0189056396484375,
-0.07183837890625,
0.07940673828125,
-0.0158233642578125,
0.0014209747314453125,
0.00647735595703125,
-0.049072265625,
0.036895751953125,
-0.007762908935546875,
-0.0285491943359375,
-0.018585205078125,
0.00785064697265625,
0.068359375,
-0.0298309326171875,
0.03350830078125,
-0.050689697265625,
0.02508544921875,
0.012359619140625,
-0.015838623046875,
0.0263214111328125,
-0.004077911376953125,
0.01218414306640625,
0.00220489501953125,
0.0245513916015625,
-0.05767822265625,
0.00981903076171875,
0.030792236328125,
-0.06658935546875,
0.00911712646484375,
-0.056121826171875,
-0.0230255126953125,
-0.01139068603515625,
0.00469207763671875,
0.050872802734375,
0.0438232421875,
-0.0352783203125,
0.0155029296875,
0.03948974609375,
-0.0195159912109375,
0.053558349609375,
0.016204833984375,
-0.028961181640625,
-0.03466796875,
0.0550537109375,
0.00922393798828125,
0.0103759765625,
0.0095672607421875,
-0.0030994415283203125,
-0.025665283203125,
-0.0291290283203125,
-0.04058837890625,
0.01552581787109375,
-0.005794525146484375,
-0.025970458984375,
-0.05419921875,
-0.03497314453125,
-0.0255584716796875,
0.00997161865234375,
-0.0275421142578125,
-0.0156097412109375,
-0.052703857421875,
-0.0109405517578125,
0.038787841796875,
0.0675048828125,
0.0119171142578125,
0.0728759765625,
-0.0435791015625,
0.0302734375,
0.015716552734375,
0.0533447265625,
-0.0303192138671875,
-0.0648193359375,
-0.03216552734375,
0.0079193115234375,
-0.0233001708984375,
-0.052825927734375,
0.015655517578125,
0.0021076202392578125,
0.03265380859375,
0.01812744140625,
-0.033447265625,
0.0308685302734375,
-0.04376220703125,
0.0704345703125,
0.034759521484375,
-0.06500244140625,
0.0280914306640625,
-0.04974365234375,
0.0343017578125,
0.03900146484375,
0.0631103515625,
-0.0157012939453125,
0.024871826171875,
-0.060272216796875,
-0.032196044921875,
0.0648193359375,
0.0304718017578125,
0.009185791015625,
0.0030689239501953125,
0.056610107421875,
0.0014104843139648438,
0.0159759521484375,
-0.07855224609375,
0.00550079345703125,
-0.036346435546875,
-0.0233154296875,
-0.005573272705078125,
-0.036224365234375,
0.0015001296997070312,
-0.0458984375,
0.051605224609375,
-0.019866943359375,
0.060791015625,
-0.0015840530395507812,
-0.00617218017578125,
0.01361846923828125,
-0.023773193359375,
0.05120849609375,
0.05487060546875,
-0.03192138671875,
-0.0178070068359375,
0.024169921875,
-0.05548095703125,
0.0108795166015625,
0.03460693359375,
0.0276947021484375,
-0.0019350051879882812,
0.0157928466796875,
0.08038330078125,
0.01132965087890625,
-0.02032470703125,
0.0611572265625,
-0.0433349609375,
-0.0212249755859375,
-0.025146484375,
-0.00897979736328125,
0.0250091552734375,
-0.0130157470703125,
-0.00461578369140625,
-0.01873779296875,
-0.0118408203125,
-0.02862548828125,
0.03253173828125,
0.01885986328125,
-0.035064697265625,
-0.036590576171875,
0.060272216796875,
0.01531219482421875,
-0.0196990966796875,
0.043731689453125,
-0.0137176513671875,
-0.032867431640625,
0.0311279296875,
0.0312347412109375,
0.04656982421875,
-0.042816162109375,
0.01007080078125,
0.039825439453125,
-0.007801055908203125,
0.0008373260498046875,
0.038665771484375,
0.00916290283203125,
-0.03643798828125,
-0.0211639404296875,
-0.06842041015625,
-0.030517578125,
0.00896453857421875,
-0.056182861328125,
0.041046142578125,
-0.045989990234375,
-0.0293121337890625,
0.014892578125,
0.031005859375,
-0.050628662109375,
0.046966552734375,
0.003253936767578125,
0.08453369140625,
-0.05877685546875,
0.072509765625,
0.042938232421875,
-0.02569580078125,
-0.08978271484375,
-0.0211334228515625,
-0.0200958251953125,
-0.06353759765625,
0.03863525390625,
0.028076171875,
0.0017004013061523438,
0.0016937255859375,
-0.0782470703125,
-0.037811279296875,
0.09027099609375,
0.044830322265625,
-0.059814453125,
-0.00577545166015625,
-0.002716064453125,
0.035064697265625,
-0.050445556640625,
0.0216522216796875,
0.0438232421875,
0.006191253662109375,
0.004276275634765625,
-0.0645751953125,
-0.0121612548828125,
-0.0249176025390625,
-0.010711669921875,
0.01125335693359375,
-0.0584716796875,
0.08587646484375,
0.006023406982421875,
-0.0019626617431640625,
0.024017333984375,
0.04754638671875,
0.0283050537109375,
0.00594329833984375,
0.045257568359375,
0.060272216796875,
0.049591064453125,
-0.016448974609375,
0.07855224609375,
-0.0300140380859375,
0.0277557373046875,
0.059722900390625,
-0.0026416778564453125,
0.0212554931640625,
0.0233154296875,
0.005218505859375,
0.031494140625,
0.0927734375,
-0.004154205322265625,
0.03363037109375,
0.0201263427734375,
-0.040924072265625,
-0.00786590576171875,
-0.0013580322265625,
-0.038909912109375,
0.0180816650390625,
0.01519775390625,
-0.034454345703125,
0.0074462890625,
0.01284027099609375,
0.0265655517578125,
-0.0233001708984375,
-0.0220489501953125,
0.0675048828125,
0.0068817138671875,
-0.041717529296875,
0.020294189453125,
0.0095367431640625,
0.0384521484375,
-0.053741455078125,
0.0233001708984375,
-0.01334381103515625,
0.01403045654296875,
0.019012451171875,
-0.052154541015625,
-0.0028820037841796875,
0.00426483154296875,
-0.019439697265625,
-0.03265380859375,
0.052459716796875,
-0.03515625,
-0.0509033203125,
0.0234832763671875,
0.0174407958984375,
0.0304107666015625,
-0.020904541015625,
-0.0753173828125,
0.0101470947265625,
0.0166015625,
-0.033416748046875,
0.021759033203125,
0.04864501953125,
0.02142333984375,
0.038360595703125,
0.02978515625,
-0.01035308837890625,
0.002635955810546875,
0.002017974853515625,
0.08203125,
-0.06597900390625,
-0.042694091796875,
-0.04656982421875,
0.01137542724609375,
0.0081329345703125,
-0.04901123046875,
0.05059814453125,
0.05389404296875,
0.0283203125,
0.0084075927734375,
0.045501708984375,
-0.00701904296875,
0.059967041015625,
-0.04095458984375,
0.022125244140625,
-0.01328277587890625,
-0.0194549560546875,
-0.0178375244140625,
-0.07696533203125,
0.025390625,
0.03466796875,
-0.00830078125,
0.0294036865234375,
0.048553466796875,
0.04754638671875,
-0.030853271484375,
0.020294189453125,
0.00392913818359375,
-0.0020771026611328125,
0.01806640625,
0.0496826171875,
0.056121826171875,
-0.031341552734375,
0.0300140380859375,
-0.0207061767578125,
-0.0262298583984375,
0.00009059906005859375,
-0.0682373046875,
-0.08978271484375,
-0.026031494140625,
-0.00289154052734375,
-0.02264404296875,
-0.01430511474609375,
0.059906005859375,
0.078369140625,
-0.038604736328125,
-0.0002961158752441406,
0.0005936622619628906,
-0.0391845703125,
0.04022216796875,
-0.01503753662109375,
0.0304718017578125,
-0.01415252685546875,
-0.037506103515625,
0.00327301025390625,
-0.0109710693359375,
0.05133056640625,
-0.006801605224609375,
-0.0097198486328125,
-0.035919189453125,
0.0159912109375,
0.0286712646484375,
0.00171661376953125,
-0.0010728836059570312,
-0.02496337890625,
-0.0088653564453125,
-0.0233154296875,
0.01491546630859375,
0.06134033203125,
-0.04132080078125,
0.005687713623046875,
0.028564453125,
0.0310516357421875,
0.08319091796875,
0.042724609375,
0.06048583984375,
-0.07843017578125,
0.043609619140625,
0.01165771484375,
0.034881591796875,
0.03460693359375,
-0.0199127197265625,
0.02557373046875,
0.04290771484375,
-0.07354736328125,
-0.05560302734375,
0.007076263427734375,
-0.08489990234375,
-0.0072174072265625,
0.086181640625,
-0.02130126953125,
-0.0227508544921875,
0.01861572265625,
-0.05059814453125,
0.0212554931640625,
-0.0085296630859375,
0.00618743896484375,
0.0587158203125,
-0.017578125,
-0.01361083984375,
-0.03436279296875,
0.06549072265625,
0.012359619140625,
-0.06951904296875,
-0.026214599609375,
0.0252532958984375,
0.052886962890625,
-0.006999969482421875,
0.0743408203125,
-0.0026836395263671875,
0.0101776123046875,
0.040283203125,
0.021728515625,
0.006168365478515625,
-0.02740478515625,
-0.0030727386474609375,
0.010833740234375,
-0.0084228515625,
-0.032684326171875
]
] |
pankajmathur/orca_mini_3b | 2023-07-13T06:30:28.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:psmathur/alpaca_orca",
"dataset:psmathur/dolly-v2_orca",
"dataset:psmathur/WizardLM_Orca",
"arxiv:2306.02707",
"license:cc-by-nc-sa-4.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | pankajmathur | null | null | pankajmathur/orca_mini_3b | 129 | 7,592 | transformers | 2023-06-22T23:13:17 | ---
license: cc-by-nc-sa-4.0
language:
- en
library_name: transformers
datasets:
- psmathur/alpaca_orca
- psmathur/dolly-v2_orca
- psmathur/WizardLM_Orca
pipeline_tag: text-generation
---
# orca_mini_3b
Use orca-mini-3b on Free Google Colab with T4 GPU :)
<a target="_blank" href="https://colab.research.google.com/#fileId=https://huggingface.co/psmathur/orca_mini_3b/blob/main/orca_mini_3b_T4_GPU.ipynb">
<img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
</a>
An [OpenLLaMa-3B model](https://github.com/openlm-research/open_llama) model trained on explain tuned datasets, created using Instructions and Input from WizardLM, Alpaca & Dolly-V2 datasets and applying Orca Research Paper dataset construction approaches.
# Dataset
We build explain tuned [WizardLM dataset ~70K](https://github.com/nlpxucan/WizardLM), [Alpaca dataset ~52K](https://crfm.stanford.edu/2023/03/13/alpaca.html) & [Dolly-V2 dataset ~15K](https://github.com/databrickslabs/dolly) created using approaches from [Orca Research Paper](https://arxiv.org/abs/2306.02707).
We leverage all of the 15 system instructions provided in Orca Research Paper. to generate custom datasets, in contrast to vanilla instruction tuning approaches used by original datasets.
This helps student model aka this model to learn ***thought*** process from teacher model, which is ChatGPT (gpt-3.5-turbo-0301 version).
Please see below example usage how the **System** prompt is added before each **instruction**.
# Training
The training configurations are provided in the table below.
The training takes on 8x A100(80G) GPUs and lasts for around 4 Hours for cost of $48 using [Lambda Labs](https://lambdalabs.com)
We used DeepSpeed with fully sharded data parallelism, also know as [ZeRO stage 3](https://engineering.fb.com/2021/07/15/open-source/fsdp/) by writing our own fine tunning scripts plus leveraging some of the model training code provided by amazing [OpenAlpaca repo](https://github.com/yxuansu/OpenAlpaca)
Here are some of params used during training:
|||
|:-------------:|:-------------:|
|*batch_size*|64|
|*train_micro_batch_size_per_gpu*|4|
|*gradient_accumulation_steps*|2|
|*Learning rate*|2e-5|
|*Max length*|1024|
|*Epochs*|3|
|*Optimizer*|AdamW|
# Example Usage
Below shows an example on how to use this model
```python
import torch
from transformers import LlamaForCausalLM, LlamaTokenizer
# Hugging Face model_path
model_path = 'psmathur/orca_mini_3b'
tokenizer = LlamaTokenizer.from_pretrained(model_path)
model = LlamaForCausalLM.from_pretrained(
model_path, torch_dtype=torch.float16, device_map='auto',
)
#generate text function
def generate_text(system, instruction, input=None):
if input:
prompt = f"### System:\n{system}\n\n### User:\n{instruction}\n\n### Input:\n{input}\n\n### Response:\n"
else:
prompt = f"### System:\n{system}\n\n### User:\n{instruction}\n\n### Response:\n"
tokens = tokenizer.encode(prompt)
tokens = torch.LongTensor(tokens).unsqueeze(0)
tokens = tokens.to('cuda')
instance = {'input_ids': tokens,'top_p': 1.0, 'temperature':0.7, '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']
)
output = rest[0][length:]
string = tokenizer.decode(output, skip_special_tokens=True)
return f'[!] Response: {string}'
# Sample Test Instruction Used by Youtuber Sam Witteveen https://www.youtube.com/@samwitteveenai
system = 'You are an AI assistant that follows instruction extremely well. Help as much as you can.'
instruction = 'Write a letter to Sam Altman, CEO of OpenAI, requesting him to convert GPT4 a private model by OpenAI to an open source project'
print(generate_text(system, instruction))
```
```
[!] Response:
Dear Sam Altman,
I am writing to request that you convert the GPT4 private model developed by OpenAI to an open source project. As a user of OpenAI, I have been waiting for the day when I can use the advanced natural language processing capabilities of GPT4 in a more open and accessible way.
While OpenAI has made significant progress in developing AI applications, it has primarily focused on building private models that are not accessible to the general public. However, with the recent release of GPT-3, there is a growing demand for more open and accessible AI tools.
Converting GPT4 to an open source project would allow for greater transparency, collaboration, and innovation. It would also help to build trust in the technology and ensure that it is used ethically and responsibly.
I urge you to consider converting GPT4 to an open source project. This would be a significant contribution to the AI community and would help to create a more open and accessible future.
Thank you for your consideration.
Sincerely,
[Your Name]
```
**P.S. I am #opentowork and #collaboration, if you can help, please reach out to me at www.linkedin.com/in/pankajam**
Next Goals:
1) Try more data like actually using FLAN-v2, just like Orka Research Paper (I am open for suggestions)
2) Provide more options for Text generation UI. (may be https://github.com/oobabooga/text-generation-webui)
3) Provide 4bit GGML/GPTQ quantized model (may be [TheBloke](https://huggingface.co/TheBloke) can help here)
Limitations & Biases:
This model can produce factually incorrect output, and should not be relied on to produce factually accurate information.
This model was trained on various public datasets. While great efforts have been taken to clean the pretraining data, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
Disclaimer:
The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model.
Please cosult an attorney before using this model for commercial purposes.
Citiation:
If you found wizardlm_alpaca_dolly_orca_open_llama_3b useful in your research or applications, please kindly cite using the following BibTeX:
```
@misc{orca_mini_3b,
author = {Pankaj Mathur},
title = {wizardlm_alpaca_dolly_orca_open_llama_3b: An explain tuned OpenLLaMA-3b model on custom wizardlm, alpaca, & dolly datasets},
year = {2023},
publisher = {GitHub, HuggingFace},
journal = {GitHub repository, HuggingFace repository},
howpublished = {\url{https://github.com/pankajarm/wizardlm_alpaca_dolly_orca_open_llama_3b}, \url{https://https://huggingface.co/psmathur/wizardlm_alpaca_dolly_orca_open_llama_3b}},
}
```
```
@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{openlm2023openllama,
author = {Xinyang Geng and Hao Liu},
title = {OpenLLaMA: An Open Reproduction of LLaMA},
month = May,
year = 2023,
url = {https://github.com/openlm-research/open_llama}
}
```
```
@misc{openalpaca,
author = {Yixuan Su and Tian Lan and Deng Cai},
title = {OpenAlpaca: A Fully Open-Source Instruction-Following Model Based On OpenLLaMA},
year = {2023},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/yxuansu/OpenAlpaca}},
}
```
```
@misc{alpaca,
author = {Rohan Taori and Ishaan Gulrajani and Tianyi Zhang and Yann Dubois and Xuechen Li and Carlos Guestrin and Percy Liang and Tatsunori B. Hashimoto },
title = {Stanford Alpaca: An Instruction-following LLaMA model},
year = {2023},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/tatsu-lab/stanford_alpaca}},
}
``` | 8,100 | [
[
-0.0233001708984375,
-0.06646728515625,
0.024932861328125,
0.00525665283203125,
-0.0156707763671875,
-0.0250396728515625,
-0.016326904296875,
-0.04486083984375,
-0.0028133392333984375,
0.024658203125,
-0.03619384765625,
-0.04925537109375,
-0.0254058837890625,
-0.0013799667358398438,
-0.0101318359375,
0.09686279296875,
-0.026763916015625,
-0.006107330322265625,
0.01227569580078125,
-0.0092620849609375,
-0.024169921875,
-0.0273284912109375,
-0.05963134765625,
-0.0172576904296875,
0.03009033203125,
0.01678466796875,
0.0460205078125,
0.04693603515625,
0.020263671875,
0.021514892578125,
-0.0008382797241210938,
0.0138397216796875,
-0.03704833984375,
-0.02313232421875,
0.009429931640625,
-0.035186767578125,
-0.05303955078125,
0.0174407958984375,
0.032318115234375,
0.0159454345703125,
-0.00972747802734375,
0.0238800048828125,
0.0121002197265625,
0.0210418701171875,
-0.041473388671875,
0.042266845703125,
-0.032073974609375,
-0.00044989585876464844,
-0.0281219482421875,
-0.0015077590942382812,
-0.0163116455078125,
-0.04779052734375,
0.0029964447021484375,
-0.068115234375,
0.0284576416015625,
-0.01107025146484375,
0.08587646484375,
0.01389312744140625,
-0.0109710693359375,
-0.0286712646484375,
-0.0491943359375,
0.054046630859375,
-0.07000732421875,
0.00984954833984375,
0.0224456787109375,
0.0225830078125,
-0.010955810546875,
-0.062225341796875,
-0.061309814453125,
-0.01473236083984375,
-0.003437042236328125,
0.017120361328125,
-0.00254058837890625,
-0.01239776611328125,
0.0167388916015625,
0.036834716796875,
-0.046478271484375,
-0.0026092529296875,
-0.043609619140625,
-0.01015472412109375,
0.0335693359375,
0.0069427490234375,
0.0176544189453125,
-0.004039764404296875,
-0.02764892578125,
-0.025909423828125,
-0.055694580078125,
0.0173492431640625,
0.0382080078125,
0.0280914306640625,
-0.030853271484375,
0.048675537109375,
-0.003910064697265625,
0.044647216796875,
-0.006011962890625,
-0.0229034423828125,
0.03875732421875,
-0.02423095703125,
-0.032440185546875,
-0.004383087158203125,
0.07196044921875,
0.00484466552734375,
0.002689361572265625,
0.00408935546875,
-0.0032672882080078125,
0.0042266845703125,
0.0004429817199707031,
-0.0657958984375,
-0.0176849365234375,
0.01129150390625,
-0.0268402099609375,
-0.0233612060546875,
-0.0057525634765625,
-0.061431884765625,
-0.00969696044921875,
-0.005664825439453125,
0.03692626953125,
-0.041595458984375,
-0.0222625732421875,
0.020263671875,
0.00920867919921875,
0.04052734375,
0.0218963623046875,
-0.07977294921875,
0.0141448974609375,
0.036102294921875,
0.07733154296875,
0.01461029052734375,
-0.0264739990234375,
-0.0187835693359375,
0.01702880859375,
-0.0173492431640625,
0.034759521484375,
-0.02020263671875,
-0.02667236328125,
-0.0181427001953125,
0.00537109375,
-0.01319122314453125,
-0.0190887451171875,
0.04486083984375,
-0.035675048828125,
0.0364990234375,
-0.0164031982421875,
-0.01540374755859375,
-0.03326416015625,
0.00823211669921875,
-0.047760009765625,
0.074951171875,
0.007770538330078125,
-0.06243896484375,
0.01483154296875,
-0.0767822265625,
-0.006618499755859375,
-0.01111602783203125,
-0.0075225830078125,
-0.0484619140625,
-0.0217437744140625,
0.04022216796875,
0.020660400390625,
-0.0273284912109375,
0.0121917724609375,
-0.02239990234375,
-0.0166778564453125,
-0.002655029296875,
-0.0240325927734375,
0.08526611328125,
0.021148681640625,
-0.04156494140625,
0.0186767578125,
-0.0537109375,
-0.006710052490234375,
0.0318603515625,
-0.03167724609375,
-0.005374908447265625,
-0.017120361328125,
-0.01103973388671875,
-0.0036106109619140625,
0.03009033203125,
-0.04833984375,
0.0297393798828125,
-0.035430908203125,
0.0526123046875,
0.0604248046875,
-0.0110015869140625,
0.0239715576171875,
-0.0174713134765625,
0.03021240234375,
-0.00847625732421875,
0.031402587890625,
-0.00223541259765625,
-0.06939697265625,
-0.0687255859375,
-0.024169921875,
0.0147705078125,
0.031585693359375,
-0.05474853515625,
0.0235137939453125,
-0.014404296875,
-0.048248291015625,
-0.049896240234375,
-0.00502777099609375,
0.0262908935546875,
0.06097412109375,
0.047088623046875,
-0.02069091796875,
-0.030517578125,
-0.045989990234375,
0.006175994873046875,
-0.01324462890625,
-0.01047515869140625,
0.0204010009765625,
0.053131103515625,
-0.00875091552734375,
0.071533203125,
-0.04791259765625,
-0.03350830078125,
-0.007724761962890625,
0.0123291015625,
0.024200439453125,
0.053558349609375,
0.04962158203125,
-0.0382080078125,
-0.0291595458984375,
0.0017709732055664062,
-0.06793212890625,
0.01120758056640625,
0.009979248046875,
-0.019989013671875,
0.0259246826171875,
0.010223388671875,
-0.0625,
0.05108642578125,
0.037322998046875,
-0.025787353515625,
0.0347900390625,
-0.014404296875,
0.00193023681640625,
-0.07012939453125,
0.02142333984375,
-0.00605010986328125,
-0.0021228790283203125,
-0.03204345703125,
0.0159149169921875,
-0.0011281967163085938,
-0.01558685302734375,
-0.034698486328125,
0.04345703125,
-0.0313720703125,
-0.001556396484375,
-0.0033111572265625,
-0.0007038116455078125,
-0.001922607421875,
0.058746337890625,
-0.00018465518951416016,
0.06378173828125,
0.039703369140625,
-0.034698486328125,
0.0219879150390625,
0.03271484375,
-0.025115966796875,
0.00719451904296875,
-0.06561279296875,
0.03021240234375,
0.0079803466796875,
0.036224365234375,
-0.05462646484375,
-0.02032470703125,
0.06353759765625,
-0.0380859375,
0.0196685791015625,
-0.0010614395141601562,
-0.035308837890625,
-0.028717041015625,
-0.0291900634765625,
0.02899169921875,
0.04229736328125,
-0.05419921875,
0.04498291015625,
0.01214599609375,
0.003814697265625,
-0.038665771484375,
-0.051177978515625,
-0.0222625732421875,
-0.017059326171875,
-0.0545654296875,
0.031524658203125,
-0.0127716064453125,
0.01091766357421875,
0.0034809112548828125,
0.00034117698669433594,
0.004238128662109375,
-0.0110015869140625,
0.018035888671875,
0.036407470703125,
-0.027679443359375,
-0.00988006591796875,
-0.005542755126953125,
-0.007411956787109375,
-0.0033779144287109375,
-0.0272064208984375,
0.045989990234375,
-0.0302276611328125,
-0.016754150390625,
-0.04156494140625,
0.004245758056640625,
0.02276611328125,
-0.0294952392578125,
0.07037353515625,
0.06304931640625,
-0.0274200439453125,
0.00567626953125,
-0.0259246826171875,
-0.0106048583984375,
-0.039215087890625,
0.0135040283203125,
-0.0260772705078125,
-0.047332763671875,
0.032928466796875,
0.0236358642578125,
0.02862548828125,
0.04541015625,
0.04571533203125,
0.0189666748046875,
0.06549072265625,
0.050750732421875,
-0.0002498626708984375,
0.031646728515625,
-0.050567626953125,
0.0092315673828125,
-0.06707763671875,
-0.038665771484375,
-0.046478271484375,
-0.0116424560546875,
-0.03887939453125,
-0.037994384765625,
0.030914306640625,
0.00891876220703125,
-0.050567626953125,
0.0262908935546875,
-0.0555419921875,
0.0181884765625,
0.04632568359375,
0.0269927978515625,
0.01465606689453125,
-0.0010013580322265625,
0.004909515380859375,
0.0159454345703125,
-0.044158935546875,
-0.05126953125,
0.09783935546875,
0.0305633544921875,
0.04620361328125,
0.00824737548828125,
0.047271728515625,
-0.0119781494140625,
0.032012939453125,
-0.032989501953125,
0.042816162109375,
0.0085296630859375,
-0.04486083984375,
-0.033172607421875,
-0.0308990478515625,
-0.08062744140625,
0.01126861572265625,
0.002002716064453125,
-0.06170654296875,
0.01261138916015625,
0.0021953582763671875,
-0.033172607421875,
0.036468505859375,
-0.05584716796875,
0.0723876953125,
-0.007678985595703125,
-0.0201263427734375,
0.007110595703125,
-0.039398193359375,
0.045196533203125,
0.001789093017578125,
0.007297515869140625,
-0.004367828369140625,
-0.0163726806640625,
0.07135009765625,
-0.05859375,
0.0670166015625,
-0.0195159912109375,
-0.0221710205078125,
0.0355224609375,
-0.0248870849609375,
0.03912353515625,
0.01314544677734375,
-0.01561737060546875,
0.036285400390625,
-0.0178070068359375,
-0.036834716796875,
-0.02313232421875,
0.06634521484375,
-0.09735107421875,
-0.0380859375,
-0.039764404296875,
-0.036285400390625,
-0.0014600753784179688,
0.0141448974609375,
0.0312347412109375,
0.019378662109375,
0.009552001953125,
-0.005985260009765625,
0.038787841796875,
-0.027862548828125,
0.040618896484375,
0.031829833984375,
-0.0169830322265625,
-0.043670654296875,
0.07098388671875,
0.00992584228515625,
0.003696441650390625,
0.01392364501953125,
0.01898193359375,
-0.0231170654296875,
-0.04217529296875,
-0.043792724609375,
0.036376953125,
-0.052001953125,
-0.0200347900390625,
-0.046844482421875,
-0.0204620361328125,
-0.0419921875,
-0.004192352294921875,
-0.0253753662109375,
-0.01947021484375,
-0.059661865234375,
-0.005496978759765625,
0.049774169921875,
0.0523681640625,
-0.0003769397735595703,
0.0254058837890625,
-0.0367431640625,
0.02801513671875,
0.032257080078125,
0.0192718505859375,
0.00815582275390625,
-0.044708251953125,
-0.01056671142578125,
0.0154266357421875,
-0.05499267578125,
-0.06329345703125,
0.03826904296875,
0.0111541748046875,
0.03173828125,
0.01338958740234375,
-0.001338958740234375,
0.062286376953125,
-0.0199432373046875,
0.07916259765625,
0.015167236328125,
-0.07293701171875,
0.04498291015625,
-0.03106689453125,
0.0160064697265625,
0.0136566162109375,
0.0384521484375,
-0.0028057098388671875,
-0.0187225341796875,
-0.0462646484375,
-0.06707763671875,
0.07940673828125,
0.036407470703125,
-0.0026531219482421875,
0.0172576904296875,
0.029449462890625,
0.0235137939453125,
0.007678985595703125,
-0.078125,
-0.0189056396484375,
-0.0455322265625,
-0.006076812744140625,
-0.011688232421875,
0.005687713623046875,
-0.013214111328125,
-0.0221710205078125,
0.07135009765625,
-0.0074920654296875,
0.042755126953125,
0.01253509521484375,
0.009765625,
-0.0036945343017578125,
0.0017147064208984375,
0.05169677734375,
0.0374755859375,
-0.0257415771484375,
-0.022216796875,
0.0113677978515625,
-0.05450439453125,
-0.0020122528076171875,
0.02166748046875,
-0.02386474609375,
-0.007213592529296875,
0.01861572265625,
0.0703125,
-0.0104827880859375,
-0.0150604248046875,
0.030731201171875,
-0.005626678466796875,
-0.0106201171875,
-0.020294189453125,
0.009735107421875,
0.00860595703125,
0.026611328125,
0.0220794677734375,
0.0096435546875,
-0.004199981689453125,
-0.040985107421875,
-0.02630615234375,
0.0166168212890625,
0.0032806396484375,
-0.0333251953125,
0.07806396484375,
0.01129150390625,
-0.01444244384765625,
0.0477294921875,
-0.00788116455078125,
-0.0202178955078125,
0.053680419921875,
0.040435791015625,
0.0643310546875,
-0.014556884765625,
0.011383056640625,
0.04412841796875,
0.023101806640625,
-0.00391387939453125,
0.026519775390625,
0.00299072265625,
-0.0301666259765625,
-0.018768310546875,
-0.049774169921875,
-0.0213165283203125,
0.034149169921875,
-0.03839111328125,
0.0408935546875,
-0.042449951171875,
-0.005329132080078125,
-0.00391387939453125,
0.005733489990234375,
-0.060150146484375,
0.01035308837890625,
0.00829315185546875,
0.061126708984375,
-0.047332763671875,
0.07879638671875,
0.036834716796875,
-0.06427001953125,
-0.0814208984375,
-0.006023406982421875,
-0.01126861572265625,
-0.06884765625,
0.03662109375,
0.016754150390625,
-0.0025806427001953125,
0.00020325183868408203,
-0.05670166015625,
-0.0692138671875,
0.10626220703125,
0.042236328125,
-0.026885986328125,
-0.02099609375,
0.0068511962890625,
0.036224365234375,
-0.0218353271484375,
0.045166015625,
0.039398193359375,
0.033416748046875,
0.008270263671875,
-0.07806396484375,
0.0194244384765625,
-0.01418304443359375,
0.00560760498046875,
0.0025272369384765625,
-0.06988525390625,
0.099853515625,
-0.02117919921875,
-0.0121307373046875,
0.02227783203125,
0.06365966796875,
0.0313720703125,
0.0248870849609375,
0.019287109375,
0.044677734375,
0.054962158203125,
-0.0096435546875,
0.07110595703125,
-0.01067352294921875,
0.04461669921875,
0.07135009765625,
-0.0025463104248046875,
0.0579833984375,
0.0176239013671875,
-0.021240234375,
0.05126953125,
0.05670166015625,
-0.003055572509765625,
0.05242919921875,
0.0038852691650390625,
-0.01143646240234375,
0.01702880859375,
0.01629638671875,
-0.06402587890625,
0.025360107421875,
0.0262298583984375,
-0.022491455078125,
-0.0184173583984375,
0.01244354248046875,
0.0181427001953125,
-0.025787353515625,
-0.0177459716796875,
0.040771484375,
-0.0027294158935546875,
-0.041259765625,
0.08343505859375,
0.00848388671875,
0.05908203125,
-0.0633544921875,
-0.0061798095703125,
-0.01800537109375,
0.01230621337890625,
-0.027618408203125,
-0.039215087890625,
0.01120758056640625,
0.01238250732421875,
0.00421905517578125,
-0.004291534423828125,
0.0291290283203125,
-0.021514892578125,
-0.0276031494140625,
0.0106658935546875,
0.00945281982421875,
0.03204345703125,
0.01082611083984375,
-0.05889892578125,
0.020111083984375,
0.01163482666015625,
-0.04266357421875,
0.0247955322265625,
0.0286102294921875,
0.00405120849609375,
0.034515380859375,
0.056671142578125,
-0.002246856689453125,
0.01459503173828125,
-0.021514892578125,
0.0836181640625,
-0.037353515625,
-0.0321044921875,
-0.05517578125,
0.032379150390625,
0.00798797607421875,
-0.0386962890625,
0.06048583984375,
0.0455322265625,
0.07818603515625,
-0.01519775390625,
0.06365966796875,
-0.0225830078125,
0.018585205078125,
-0.04449462890625,
0.054962158203125,
-0.03619384765625,
0.02581787109375,
-0.022308349609375,
-0.068359375,
-0.002056121826171875,
0.059295654296875,
-0.0372314453125,
0.01299285888671875,
0.0462646484375,
0.067626953125,
-0.01259613037109375,
0.020965576171875,
-0.0005393028259277344,
0.0168914794921875,
0.044525146484375,
0.057403564453125,
0.03424072265625,
-0.047149658203125,
0.06451416015625,
-0.0333251953125,
-0.032196044921875,
-0.006130218505859375,
-0.0631103515625,
-0.06451416015625,
-0.034515380859375,
-0.02581787109375,
-0.0301513671875,
0.0014638900756835938,
0.0526123046875,
0.047332763671875,
-0.055755615234375,
-0.0218963623046875,
-0.018463134765625,
-0.00415802001953125,
-0.0206451416015625,
-0.0171661376953125,
0.05126953125,
-0.011383056640625,
-0.07196044921875,
0.01904296875,
-0.00891876220703125,
0.0203094482421875,
-0.0252532958984375,
-0.0167999267578125,
-0.01471710205078125,
0.0019817352294921875,
0.0272369384765625,
0.040802001953125,
-0.057586669921875,
-0.01605224609375,
-0.0138397216796875,
-0.010986328125,
0.01551055908203125,
0.03631591796875,
-0.0638427734375,
0.027618408203125,
0.02178955078125,
0.01849365234375,
0.059844970703125,
-0.0201416015625,
0.0208892822265625,
-0.0303497314453125,
0.01898193359375,
0.01261138916015625,
0.0219879150390625,
0.021087646484375,
-0.02825927734375,
0.054351806640625,
0.0186004638671875,
-0.050994873046875,
-0.05865478515625,
-0.003814697265625,
-0.07672119140625,
-0.00858306884765625,
0.08172607421875,
-0.03741455078125,
-0.036224365234375,
0.005458831787109375,
-0.03350830078125,
0.038909912109375,
-0.044525146484375,
0.0587158203125,
0.023956298828125,
-0.0139312744140625,
0.002162933349609375,
-0.04931640625,
0.033447265625,
0.00037598609924316406,
-0.0723876953125,
-0.013427734375,
0.0200347900390625,
0.0266571044921875,
0.0215911865234375,
0.05548095703125,
-0.01009368896484375,
0.01421356201171875,
0.004669189453125,
0.02069091796875,
-0.029388427734375,
-0.0047760009765625,
-0.01499176025390625,
0.01202392578125,
-0.0196685791015625,
-0.045806884765625
]
] |
totally-not-an-llm/EverythingLM-13b-V3-16k | 2023-09-23T23:30:34.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:totally-not-an-llm/EverythingLM-data-V3",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | totally-not-an-llm | null | null | totally-not-an-llm/EverythingLM-13b-V3-16k | 4 | 7,580 | transformers | 2023-09-23T01:18:48 | ---
license: llama2
datasets:
- totally-not-an-llm/EverythingLM-data-V3
---
# EverythingLM-13b-V3-16k
Introducing EverythingLM, a llama-2 based, general-purpose 13b model with 16k context thanks to LlongMa. The model is trained on the EverythingLM-V3 dataset, more info can be found on the dataset page.
The model is completely uncensored.
Despite being "uncensored", the base model might be resistant; you might have to prompt-engineer certain prompts.
### Quants (Thanks TheBloke!):
https://huggingface.co/TheBloke/EverythingLM-13B-V3-16K-GGUF
https://huggingface.co/TheBloke/EverythingLM-13B-V3-16K-GPTQ
https://huggingface.co/TheBloke/EverythingLM-13B-V3-16K-AWQ
### Notable features:
- Automatically triggered CoT reasoning.
- Verbose and detailed replies.
- Creative stories.
- Good prompt understanding.
### Differences from V2:
- Much more uncensored.
- Actual roleplaying ability now!
- General all around improvements thanks to the new dataset. Check out the dataset for more info.
### Prompt format (Alpaca-chat):
```
USER: <prompt>
ASSISTANT:
```
### Future plans:
- Highest priority right now is V3.1 with more optimized training and iterative dataset improvements based on testing.
### Note:
Through testing V2, I realized some alignment data had leaked in, causing the model to be less cooperative then intended. This model should do much better due to stricter filetering. | 1,404 | [
[
-0.009063720703125,
-0.06768798828125,
0.039154052734375,
0.046905517578125,
-0.044158935546875,
-0.0011119842529296875,
0.0219573974609375,
-0.051239013671875,
0.0153656005859375,
0.05859375,
-0.05810546875,
-0.05322265625,
-0.04656982421875,
-0.006191253662109375,
-0.025787353515625,
0.0848388671875,
0.0222930908203125,
0.00435638427734375,
0.0023956298828125,
0.0036449432373046875,
-0.042083740234375,
-0.0467529296875,
-0.03717041015625,
-0.024688720703125,
0.03924560546875,
0.02301025390625,
0.05389404296875,
0.0660400390625,
0.01245880126953125,
0.01519012451171875,
-0.01641845703125,
0.0216064453125,
-0.054290771484375,
0.0028400421142578125,
-0.0034236907958984375,
-0.0063323974609375,
-0.03997802734375,
-0.00800323486328125,
0.0280303955078125,
0.01531219482421875,
-0.02960205078125,
0.02630615234375,
-0.007213592529296875,
0.0306396484375,
-0.04437255859375,
0.0255889892578125,
-0.04559326171875,
-0.004486083984375,
-0.0225067138671875,
0.016448974609375,
-0.03118896484375,
-0.0036296844482421875,
-0.020477294921875,
-0.0614013671875,
-0.0055084228515625,
0.034698486328125,
0.07489013671875,
0.021331787109375,
-0.053253173828125,
-0.014068603515625,
-0.033050537109375,
0.0526123046875,
-0.047393798828125,
0.015960693359375,
0.05157470703125,
0.0247802734375,
-0.0111236572265625,
-0.052947998046875,
-0.020263671875,
-0.0241546630859375,
0.01263427734375,
0.007305145263671875,
-0.03460693359375,
0.011810302734375,
0.00638580322265625,
0.01531219482421875,
-0.029052734375,
0.020263671875,
-0.03277587890625,
-0.0279998779296875,
0.05718994140625,
0.03106689453125,
0.0032024383544921875,
-0.0208892822265625,
-0.038421630859375,
0.007518768310546875,
-0.057281494140625,
-0.0036869049072265625,
0.031341552734375,
0.005535125732421875,
-0.030548095703125,
0.058990478515625,
0.0007781982421875,
0.06011962890625,
0.00043964385986328125,
-0.002666473388671875,
0.0024204254150390625,
-0.021331787109375,
-0.0380859375,
-0.01297760009765625,
0.046600341796875,
0.04248046875,
-0.01337432861328125,
0.00328826904296875,
-0.0121917724609375,
-0.004032135009765625,
0.0173797607421875,
-0.0477294921875,
-0.0092926025390625,
0.01003265380859375,
-0.0526123046875,
-0.045074462890625,
-0.0021190643310546875,
-0.0264129638671875,
-0.05328369140625,
-0.0220489501953125,
0.029083251953125,
-0.027740478515625,
-0.0284423828125,
0.0038909912109375,
0.0191650390625,
0.04229736328125,
0.040740966796875,
-0.043304443359375,
0.029876708984375,
0.0587158203125,
0.052490234375,
-0.01074981689453125,
-0.01284027099609375,
-0.026641845703125,
0.006763458251953125,
-0.03631591796875,
0.04595947265625,
-0.0212860107421875,
-0.036712646484375,
-0.0146942138671875,
0.025726318359375,
0.0125885009765625,
-0.0260009765625,
0.041778564453125,
-0.035888671875,
0.0059356689453125,
-0.01424407958984375,
-0.046630859375,
-0.017974853515625,
0.010589599609375,
-0.052001953125,
0.08135986328125,
0.01490020751953125,
-0.0279693603515625,
-0.0109405517578125,
-0.062347412109375,
-0.005580902099609375,
-0.007335662841796875,
0.015533447265625,
-0.0267486572265625,
-0.01210784912109375,
-0.0026607513427734375,
0.030975341796875,
-0.043060302734375,
0.03076171875,
-0.0254058837890625,
-0.033050537109375,
0.0200042724609375,
-0.025848388671875,
0.0767822265625,
0.02813720703125,
-0.01432037353515625,
0.0269775390625,
-0.0638427734375,
-0.0127716064453125,
0.0191650390625,
-0.00229644775390625,
0.02294921875,
-0.0311279296875,
0.0274505615234375,
0.005245208740234375,
0.034515380859375,
-0.041107177734375,
0.011749267578125,
0.00464630126953125,
0.0239105224609375,
0.06658935546875,
0.007495880126953125,
0.0296630859375,
-0.0411376953125,
0.05206298828125,
-0.0021076202392578125,
0.054840087890625,
0.00907135009765625,
-0.074951171875,
-0.0372314453125,
-0.0318603515625,
0.00665283203125,
0.048828125,
-0.039276123046875,
0.0343017578125,
0.036834716796875,
-0.061737060546875,
-0.039642333984375,
0.01258087158203125,
0.0396728515625,
0.032073974609375,
0.024322509765625,
-0.0006327629089355469,
-0.056976318359375,
-0.063720703125,
0.0284423828125,
-0.028961181640625,
-0.021087646484375,
0.0237274169921875,
0.021240234375,
-0.0280303955078125,
0.03704833984375,
-0.037139892578125,
-0.0270538330078125,
-0.037139892578125,
-0.013916015625,
0.00786590576171875,
0.02508544921875,
0.040313720703125,
-0.03875732421875,
-0.006855010986328125,
-0.0112457275390625,
-0.051422119140625,
-0.01422119140625,
0.0022983551025390625,
-0.045501708984375,
0.01751708984375,
0.0245208740234375,
-0.05096435546875,
0.03973388671875,
0.047088623046875,
-0.043670654296875,
0.046142578125,
-0.00409698486328125,
-0.0208587646484375,
-0.088134765625,
-0.0018596649169921875,
-0.005771636962890625,
-0.02008056640625,
-0.03070068359375,
0.0099945068359375,
0.00714111328125,
-0.007274627685546875,
-0.048553466796875,
0.04888916015625,
-0.03143310546875,
-0.0141448974609375,
-0.029388427734375,
-0.0031986236572265625,
0.016754150390625,
0.038299560546875,
-0.010467529296875,
0.037261962890625,
0.0562744140625,
-0.042205810546875,
0.056060791015625,
0.055633544921875,
-0.022552490234375,
0.00983428955078125,
-0.06365966796875,
0.019256591796875,
-0.005634307861328125,
0.042510986328125,
-0.06390380859375,
-0.027587890625,
0.037628173828125,
-0.043304443359375,
0.01171112060546875,
-0.00930023193359375,
-0.034759521484375,
-0.0096282958984375,
-0.032135009765625,
0.0004200935363769531,
0.04071044921875,
-0.03619384765625,
0.0472412109375,
0.0298004150390625,
0.0169219970703125,
-0.055999755859375,
-0.056427001953125,
0.0065155029296875,
-0.0181732177734375,
-0.03570556640625,
0.0241851806640625,
-0.0141448974609375,
-0.0228729248046875,
0.002422332763671875,
-0.0010728836059570312,
-0.027587890625,
0.019775390625,
0.037139892578125,
0.0595703125,
-0.00972747802734375,
-0.0103302001953125,
-0.0140380859375,
0.0019550323486328125,
0.00249481201171875,
0.0134735107421875,
0.033203125,
-0.01678466796875,
-0.017181396484375,
-0.031585693359375,
0.0252532958984375,
0.0144195556640625,
0.005702972412109375,
0.06903076171875,
0.04656982421875,
-0.024444580078125,
0.00833892822265625,
-0.050750732421875,
0.0008749961853027344,
-0.03350830078125,
0.0217437744140625,
-0.0033931732177734375,
-0.0709228515625,
0.048828125,
0.0284881591796875,
0.00493621826171875,
0.019989013671875,
0.0399169921875,
-0.00293731689453125,
0.06817626953125,
0.052825927734375,
-0.004642486572265625,
0.0299835205078125,
-0.0034542083740234375,
-0.0002313852310180664,
-0.06695556640625,
-0.05120849609375,
-0.0172119140625,
-0.0175323486328125,
-0.03240966796875,
-0.03631591796875,
-0.003810882568359375,
0.0176544189453125,
-0.049896240234375,
0.034820556640625,
-0.03924560546875,
0.051239013671875,
0.037628173828125,
0.0280914306640625,
0.0245819091796875,
-0.02655029296875,
0.0311737060546875,
0.01531982421875,
-0.03948974609375,
-0.04156494140625,
0.09307861328125,
0.0295867919921875,
0.059173583984375,
0.0283203125,
0.0335693359375,
0.0311279296875,
0.0252227783203125,
-0.05322265625,
0.041412353515625,
0.006801605224609375,
-0.062744140625,
-0.01503753662109375,
-0.01629638671875,
-0.055877685546875,
0.00496673583984375,
-0.00714111328125,
-0.051300048828125,
0.00827789306640625,
-0.00279998779296875,
-0.0265655517578125,
0.0275726318359375,
-0.057769775390625,
0.04656982421875,
0.000370025634765625,
-0.01503753662109375,
0.00804901123046875,
-0.073974609375,
0.04864501953125,
-0.0173492431640625,
0.0039215087890625,
-0.026641845703125,
-0.01153564453125,
0.043701171875,
-0.033416748046875,
0.0791015625,
-0.0193023681640625,
-0.0284576416015625,
0.032135009765625,
0.013427734375,
0.03436279296875,
0.02362060546875,
-0.015350341796875,
0.0240631103515625,
-0.0017871856689453125,
-0.039093017578125,
-0.01934814453125,
0.050079345703125,
-0.08685302734375,
-0.050018310546875,
-0.033599853515625,
-0.031951904296875,
-0.0083160400390625,
-0.0105438232421875,
0.033172607421875,
0.0153045654296875,
-0.01290130615234375,
0.033660888671875,
0.040740966796875,
-0.025543212890625,
0.0207061767578125,
0.0589599609375,
-0.0269927978515625,
-0.0200347900390625,
0.039642333984375,
-0.0095672607421875,
0.0162353515625,
0.0176239013671875,
-0.0012149810791015625,
-0.0491943359375,
-0.019439697265625,
-0.0399169921875,
0.0150604248046875,
-0.05169677734375,
-0.044647216796875,
-0.046875,
-0.041839599609375,
-0.0238189697265625,
-0.003948211669921875,
-0.0232696533203125,
-0.048309326171875,
-0.049072265625,
-0.0360107421875,
0.062347412109375,
0.07037353515625,
-0.031585693359375,
0.04913330078125,
-0.043182373046875,
0.02923583984375,
0.0169219970703125,
0.01100921630859375,
-0.0012950897216796875,
-0.0650634765625,
-0.0029544830322265625,
-0.0032291412353515625,
-0.044586181640625,
-0.06024169921875,
0.02215576171875,
0.0192718505859375,
0.034149169921875,
0.03411865234375,
0.017669677734375,
0.03802490234375,
-0.0298309326171875,
0.07684326171875,
0.00994873046875,
-0.07354736328125,
0.029144287109375,
-0.031219482421875,
-0.00940704345703125,
0.008087158203125,
0.01334381103515625,
-0.02410888671875,
-0.01261138916015625,
-0.05767822265625,
-0.05511474609375,
0.062042236328125,
0.02947998046875,
0.01436614990234375,
-0.00024628639221191406,
0.0198211669921875,
0.02239990234375,
0.0188446044921875,
-0.07318115234375,
-0.03411865234375,
-0.048065185546875,
-0.0085601806640625,
0.0115509033203125,
-0.031280517578125,
-0.02288818359375,
-0.021331787109375,
0.06121826171875,
0.004261016845703125,
0.032470703125,
-0.01334381103515625,
-0.001766204833984375,
-0.00470733642578125,
-0.00112152099609375,
0.03759765625,
0.059173583984375,
-0.027618408203125,
-0.01369476318359375,
0.01128387451171875,
-0.039093017578125,
0.0285186767578125,
0.01134490966796875,
0.007335662841796875,
-0.008056640625,
0.0303192138671875,
0.09295654296875,
0.020965576171875,
-0.03936767578125,
0.04620361328125,
-0.0049285888671875,
-0.023956298828125,
-0.019989013671875,
0.004268646240234375,
0.019134521484375,
0.025604248046875,
0.0217437744140625,
-0.01009368896484375,
0.00734710693359375,
-0.029541015625,
0.007335662841796875,
0.010223388671875,
-0.010345458984375,
-0.017913818359375,
0.06781005859375,
0.0152435302734375,
-0.0122833251953125,
0.0293426513671875,
-0.0200653076171875,
-0.0099334716796875,
0.060577392578125,
0.06781005859375,
0.04583740234375,
-0.02313232421875,
0.05078125,
0.0160369873046875,
0.0227203369140625,
-0.00829315185546875,
0.023681640625,
-0.005001068115234375,
-0.056182861328125,
-0.040985107421875,
-0.0325927734375,
-0.0650634765625,
0.031585693359375,
-0.0692138671875,
0.0292816162109375,
-0.06591796875,
0.01070404052734375,
-0.0230712890625,
0.01023101806640625,
-0.038238525390625,
0.015960693359375,
0.00848388671875,
0.06341552734375,
-0.05438232421875,
0.07159423828125,
0.040802001953125,
-0.0247039794921875,
-0.06573486328125,
-0.020721435546875,
0.0247039794921875,
-0.10565185546875,
0.0249176025390625,
0.007083892822265625,
0.013427734375,
0.0145416259765625,
-0.08575439453125,
-0.07635498046875,
0.10272216796875,
0.0267333984375,
-0.041961669921875,
-0.0005335807800292969,
0.000690460205078125,
0.02734375,
-0.0289306640625,
0.0235595703125,
0.0357666015625,
0.02630615234375,
0.025421142578125,
-0.05438232421875,
0.008544921875,
-0.0275726318359375,
0.00801849365234375,
-0.0204010009765625,
-0.06951904296875,
0.060577392578125,
-0.00821685791015625,
-0.0005898475646972656,
0.039093017578125,
0.043670654296875,
0.059600830078125,
0.0279998779296875,
0.0360107421875,
0.018829345703125,
0.06317138671875,
-0.007232666015625,
0.06884765625,
-0.0170745849609375,
0.0097808837890625,
0.08453369140625,
-0.040771484375,
0.03558349609375,
0.00626373291015625,
-0.0007328987121582031,
0.021392822265625,
0.079833984375,
-0.01435089111328125,
0.0477294921875,
0.0133056640625,
-0.018890380859375,
-0.0209197998046875,
-0.032196044921875,
-0.0452880859375,
0.02020263671875,
0.0164642333984375,
0.00028705596923828125,
-0.0135040283203125,
0.00913238525390625,
0.0095062255859375,
-0.0183563232421875,
-0.02166748046875,
0.062225341796875,
0.013946533203125,
0.0007390975952148438,
0.045501708984375,
0.006561279296875,
0.053131103515625,
-0.0753173828125,
0.00687408447265625,
-0.03485107421875,
-0.0091094970703125,
-0.01129913330078125,
-0.06195068359375,
0.0158538818359375,
0.0225677490234375,
-0.006061553955078125,
-0.005832672119140625,
0.05877685546875,
-0.01422882080078125,
-0.033905029296875,
0.03131103515625,
0.0291595458984375,
0.02203369140625,
-0.0211029052734375,
-0.06341552734375,
0.037445068359375,
0.01715087890625,
-0.015960693359375,
0.0213165283203125,
0.048309326171875,
-0.010101318359375,
0.0638427734375,
0.051483154296875,
-0.0006508827209472656,
0.0021228790283203125,
0.0127716064453125,
0.07684326171875,
-0.05499267578125,
-0.033050537109375,
-0.04736328125,
0.0248870849609375,
-0.004108428955078125,
-0.030181884765625,
0.053497314453125,
0.027069091796875,
0.043548583984375,
-0.00949859619140625,
0.048431396484375,
-0.008209228515625,
0.0323486328125,
-0.040130615234375,
0.0433349609375,
-0.038482666015625,
0.018035888671875,
-0.01091766357421875,
-0.07830810546875,
-0.0102996826171875,
0.04638671875,
0.0205841064453125,
-0.01708984375,
0.04278564453125,
0.0638427734375,
0.007320404052734375,
0.026031494140625,
0.03790283203125,
0.0260009765625,
0.02276611328125,
0.04010009765625,
0.08343505859375,
-0.041259765625,
0.054779052734375,
-0.02777099609375,
-0.024871826171875,
-0.019989013671875,
-0.0665283203125,
-0.06085205078125,
-0.034576416015625,
-0.01398468017578125,
-0.037567138671875,
0.0027942657470703125,
0.07818603515625,
0.048095703125,
-0.05419921875,
-0.0174713134765625,
0.0012884140014648438,
-0.007022857666015625,
0.0025959014892578125,
-0.011444091796875,
0.01629638671875,
0.016510009765625,
-0.0262298583984375,
0.049163818359375,
-0.01395416259765625,
0.025299072265625,
-0.011138916015625,
-0.01369476318359375,
-0.01190948486328125,
0.01195526123046875,
0.049957275390625,
0.0167999267578125,
-0.050994873046875,
-0.01091766357421875,
0.00955963134765625,
0.004058837890625,
-0.005908966064453125,
0.0212860107421875,
-0.02337646484375,
-0.008453369140625,
-0.00824737548828125,
0.034637451171875,
0.04498291015625,
0.0191497802734375,
0.04132080078125,
-0.0565185546875,
0.0335693359375,
0.01556396484375,
0.0232391357421875,
0.03643798828125,
-0.04852294921875,
0.0595703125,
0.0027217864990234375,
-0.05877685546875,
-0.06439208984375,
0.0118408203125,
-0.0799560546875,
-0.0201873779296875,
0.100830078125,
-0.0002880096435546875,
-0.032470703125,
-0.01375579833984375,
-0.05718994140625,
0.026611328125,
-0.047576904296875,
0.05206298828125,
0.06610107421875,
-0.011077880859375,
-0.02203369140625,
-0.039093017578125,
0.038482666015625,
0.01995849609375,
-0.0648193359375,
0.006320953369140625,
0.051300048828125,
0.0175628662109375,
-0.0160064697265625,
0.049530029296875,
-0.0160675048828125,
0.01235198974609375,
-0.0208282470703125,
-0.0025234222412109375,
-0.0313720703125,
-0.0252685546875,
-0.01568603515625,
-0.01422119140625,
-0.0113372802734375,
-0.0299224853515625
]
] |
mosaicml/mpt-7b-8k-instruct | 2023-10-30T21:53:57.000Z | [
"transformers",
"pytorch",
"mpt",
"text-generation",
"Composer",
"MosaicML",
"llm-foundry",
"custom_code",
"dataset:competition_math",
"dataset:conceptofmind/cot_submix_original/cot_gsm8k",
"dataset:knkarthick/dialogsum",
"dataset:mosaicml/dolly_hhrlhf",
"dataset:duorc",
"dataset:tau/scrolls/qasper",
"dataset:emozilla/quality",
"dataset:scrolls/summ_screen_fd",
"dataset:spider",
"arxiv:2205.14135",
"arxiv:2108.12409",
"arxiv:2010.04245",
"license:cc-by-sa-3.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | mosaicml | null | null | mosaicml/mpt-7b-8k-instruct | 23 | 7,577 | transformers | 2023-06-18T22:32:42 | ---
license: cc-by-sa-3.0
datasets:
- competition_math
- conceptofmind/cot_submix_original/cot_gsm8k
- knkarthick/dialogsum
- mosaicml/dolly_hhrlhf
- duorc
- tau/scrolls/qasper
- emozilla/quality
- scrolls/summ_screen_fd
- spider
tags:
- Composer
- MosaicML
- llm-foundry
inference: false
---
# MPT-7B-Instruct-8k
MPT-7B-Instruct-8k is a model for long-form instruction following, especially question-answering on and summarization of longer documents.
It is built by finetuning [MPT-7B-8k](https://huggingface.co/mosaicml/mpt-7b-8k) on [Dolly HHRLHF](https://huggingface.co/datasets/mosaicml/dolly_hhrlhf) derived from the [Databricks Dolly-15k](https://huggingface.co/datasets/databricks/databricks-dolly-15k) and the [Anthropic Helpful and Harmless (HH-RLHF)](https://huggingface.co/datasets/Anthropic/hh-rlhf) datasets. It is also trained on [Competition Math](https://huggingface.co/datasets/competition_math), [Duorc](https://huggingface.co/datasets/duorc), [CoT GSM8k](https://huggingface.co/datasets/conceptofmind/cot_submix_original), [Qasper](https://huggingface.co/datasets/allenai/qasper), [Quality](https://huggingface.co/datasets/emozilla/quality), [Summ Screen FD](https://huggingface.co/datasets/tau/scrolls) and [Spider](https://huggingface.co/datasets/spider).
This is the same dataset that [MPT-30B-Instruct](https://huggingface.co/mosaicml/mpt-30b-instruct) was trained on.
* License: _CC-By-SA-3.0_
This model was trained by [MosaicML](https://www.mosaicml.com) and follows a modified decoder-only transformer architecture.
## Model Date
July 18, 2023
## Model License
_CC-By-SA-3.0_
## Documentation
* [Blog post: MPT-7B-8k](https://www.mosaicml.com/blog/long-context-mpt-7b-8k)
* [Codebase (mosaicml/llm-foundry repo)](https://github.com/mosaicml/llm-foundry/)
* Questions: Feel free to contact us via the [MosaicML Community Slack](https://mosaicml.me/slack)!
## How to Use
This model is best used with the MosaicML [llm-foundry repository](https://github.com/mosaicml/llm-foundry) for training and finetuning.
```python
import transformers
model = transformers.AutoModelForCausalLM.from_pretrained(
'mosaicml/mpt-7b-instruct-8k',
trust_remote_code=True
)
```
Note: This model requires that `trust_remote_code=True` be passed to the `from_pretrained` method.
This is because we use a custom `MPT` model architecture that is not yet part of the Hugging Face `transformers` package.
`MPT` includes options for many training efficiency features such as [FlashAttention](https://arxiv.org/pdf/2205.14135.pdf), [ALiBi](https://arxiv.org/abs/2108.12409), [QK LayerNorm](https://arxiv.org/abs/2010.04245), and more.
To use the optimized [triton implementation](https://github.com/openai/triton) of FlashAttention, you can load the model on GPU (`cuda:0`) with `attn_impl='triton'` and with `bfloat16` precision:
```python
import torch
import transformers
name = 'mosaicml/mpt-7b-instruct-8k'
config = transformers.AutoConfig.from_pretrained(name, trust_remote_code=True)
config.attn_config['attn_impl'] = 'triton' # change this to use triton-based FlashAttention
config.init_device = 'cuda:0' # For fast initialization directly on GPU!
model = transformers.AutoModelForCausalLM.from_pretrained(
name,
config=config,
torch_dtype=torch.bfloat16, # Load model weights in bfloat16
trust_remote_code=True
)
```
The model was trained initially with a sequence length of 2048 with an additional pretraining stage for sequence length adapation up to 8192. However, ALiBi enables users to increase the maximum sequence length even further during finetuning and/or inference. For example:
```python
import transformers
name = 'mosaicml/mpt-7b-instruct-8k'
config = transformers.AutoConfig.from_pretrained(name, trust_remote_code=True)
config.max_seq_len = 16384 # (input + output) tokens can now be up to 16384
model = transformers.AutoModelForCausalLM.from_pretrained(
name,
config=config,
trust_remote_code=True
)
```
This model was trained with the MPT-7B-chat tokenizer which is based on the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer and includes additional ChatML tokens.
```python
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained('mosaicml/mpt-7b-8k')
```
The model can then be used, for example, within a text-generation pipeline.
Note: when running Torch modules in lower precision, it is best practice to use the [torch.autocast context manager](https://pytorch.org/docs/stable/amp.html).
```python
from transformers import pipeline
with torch.autocast('cuda', dtype=torch.bfloat16):
inputs = tokenizer('Here is a recipe for vegan banana bread:\n', return_tensors="pt").to('cuda')
outputs = model.generate(**inputs, max_new_tokens=100)
print(tokenizer.batch_decode(outputs, skip_special_tokens=True))
# or using the HF pipeline
pipe = pipeline('text-generation', model=model, tokenizer=tokenizer, device='cuda:0')
with torch.autocast('cuda', dtype=torch.bfloat16):
print(
pipe('Here is a recipe for vegan banana bread:\n',
max_new_tokens=100,
do_sample=True,
use_cache=True))
```
## Model Description
The architecture is a modification of a standard decoder-only transformer.
The model has been modified from a standard transformer in the following ways:
* It uses [FlashAttention](https://arxiv.org/pdf/2205.14135.pdf)
* It uses [ALiBi (Attention with Linear Biases)](https://arxiv.org/abs/2108.12409) and does not use positional embeddings
* It does not use biases
| Hyperparameter | Value |
|----------------|-------|
|n_parameters | 6.7B |
|n_layers | 32 |
| n_heads | 32 |
| d_model | 4096 |
| vocab size | 50432 |
| sequence length | 2048 |
## Data Mix
The model was trained on the following data mix:
| Data Source | Number of Tokens in Source | Proportion |
|-------------|----------------------------|------------|
| competition_math | 1.6 M | 3.66% |
| cot_gsm8k | 3.36 M | 7.67% |
| dialogsum | 0.1 M | 0.23% |
| dolly_hhrlhf | 5.89 M | 13.43% |
| duorc | 7.8 M | 17.80% |
| qasper | 8.72 M | 19.90% |
| quality | 11.29 M | 25.78% |
| scrolls/summ_screen_fd | 4.97 M | 11.33% |
| spider | 0.089 M | 0.20% |
### Training Configuration
This model was trained on 8 80GB A100s for about 6.3 hours using the [MosaicML Platform](https://www.mosaicml.com/platform).
The model was trained with sharded data parallelism using [FSDP](https://pytorch.org/docs/stable/fsdp.html) and used the AdamW optimizer.
## Limitations and Biases
_The following language is modified from [EleutherAI's GPT-NeoX-20B](https://huggingface.co/EleutherAI/gpt-neox-20b)_
MPT-7B-Instruct-8k can produce factually incorrect output, and should not be relied on to produce factually accurate information.
MPT-7B-Instruct-8k was trained on various public datasets.
While great efforts have been taken to clean the pretraining data, it is possible that this model could generate lewd, biased or otherwise offensive outputs.
## Acknowledgements
This model was finetuned by the MosaicML NLP team.
## Disclaimer
The license on this model does not constitute legal advice. We are not responsible for the actions of third parties who use this model. Please consult an attorney before using this model for commercial purposes.
## MosaicML Platform
If you're interested in [training](https://www.mosaicml.com/training) and [deploying](https://www.mosaicml.com/inference) your own MPT or LLMs on the MosaicML Platform, [sign up here](https://www.mosaicml.com/get-started?utm_source=huggingface&utm_medium=referral&utm_campaign=mpt-7b-8k).
## Citation
Please cite this model using the following format:
```
@online{MosaicML2023Introducing,
author = {MosaicML NLP Team},
title = {Introducing MPT-30B: Raising the bar
for open-source foundation models},
year = {2023},
url = {www.mosaicml.com/blog/mpt-30b},
note = {Accessed: 2023-06-22},
urldate = {2023-06-22}
}
``` | 8,005 | [
[
-0.034210205078125,
-0.03619384765625,
0.01236724853515625,
0.025634765625,
-0.0246429443359375,
-0.0037326812744140625,
-0.00024056434631347656,
-0.0218658447265625,
0.00885772705078125,
0.0241241455078125,
-0.0380859375,
-0.041015625,
-0.052490234375,
0.0089263916015625,
-0.0277557373046875,
0.07666015625,
-0.00015485286712646484,
0.0002294778823852539,
0.0033016204833984375,
-0.00643157958984375,
-0.0092926025390625,
-0.0308837890625,
-0.0474853515625,
-0.0251312255859375,
0.03680419921875,
0.01329803466796875,
0.058074951171875,
0.07037353515625,
0.034332275390625,
0.024871826171875,
-0.02008056640625,
0.0107879638671875,
-0.03338623046875,
-0.030975341796875,
0.0034770965576171875,
-0.0316162109375,
-0.038726806640625,
0.0159149169921875,
0.038726806640625,
0.0290374755859375,
-0.007537841796875,
0.0287322998046875,
0.002376556396484375,
0.029083251953125,
-0.034912109375,
0.0193634033203125,
-0.0278167724609375,
0.0120849609375,
-0.007015228271484375,
-0.0022411346435546875,
-0.04083251953125,
-0.020477294921875,
0.005168914794921875,
-0.038055419921875,
0.0206298828125,
-0.003971099853515625,
0.0782470703125,
0.0168914794921875,
-0.034698486328125,
-0.0006299018859863281,
-0.04052734375,
0.044769287109375,
-0.0648193359375,
0.02899169921875,
0.024993896484375,
0.015655517578125,
-0.0019197463989257812,
-0.07550048828125,
-0.053436279296875,
-0.016021728515625,
-0.0059814453125,
0.0247955322265625,
-0.0177764892578125,
-0.0006580352783203125,
0.03570556640625,
0.030059814453125,
-0.0460205078125,
-0.00849151611328125,
-0.0338134765625,
-0.0135650634765625,
0.0355224609375,
0.01113128662109375,
0.023773193359375,
-0.034027099609375,
-0.048583984375,
-0.0299072265625,
-0.04986572265625,
0.006313323974609375,
0.033905029296875,
-0.0018463134765625,
-0.03411865234375,
0.049591064453125,
-0.00222015380859375,
0.041259765625,
0.0194244384765625,
-0.00719451904296875,
0.032318115234375,
-0.02435302734375,
-0.02447509765625,
-0.00836181640625,
0.076171875,
0.026092529296875,
0.004451751708984375,
-0.00518798828125,
-0.006587982177734375,
-0.0032176971435546875,
0.005931854248046875,
-0.08056640625,
-0.03155517578125,
0.0158538818359375,
-0.033966064453125,
-0.0157623291015625,
0.0111541748046875,
-0.048583984375,
-0.0196075439453125,
-0.0153350830078125,
0.053436279296875,
-0.05712890625,
-0.0291900634765625,
0.0037479400634765625,
-0.01332855224609375,
0.0288543701171875,
0.00457763671875,
-0.06060791015625,
0.0069580078125,
0.0311737060546875,
0.07525634765625,
-0.004703521728515625,
-0.041534423828125,
-0.01085662841796875,
0.0057830810546875,
-0.0022258758544921875,
0.039947509765625,
-0.015106201171875,
-0.01611328125,
-0.03326416015625,
0.017913818359375,
-0.026153564453125,
-0.03253173828125,
0.0218505859375,
-0.0251617431640625,
0.03369140625,
-0.01554107666015625,
-0.036895751953125,
-0.01971435546875,
0.01029205322265625,
-0.0474853515625,
0.076171875,
0.0323486328125,
-0.070068359375,
0.0216064453125,
-0.052520751953125,
-0.010986328125,
-0.01091766357421875,
0.0037326812744140625,
-0.0550537109375,
-0.0155487060546875,
0.0270538330078125,
0.03265380859375,
-0.0232086181640625,
0.01523590087890625,
-0.0210113525390625,
-0.035308837890625,
0.0146026611328125,
-0.03509521484375,
0.0771484375,
0.0206756591796875,
-0.0556640625,
0.01090240478515625,
-0.057830810546875,
-0.01110076904296875,
0.022705078125,
-0.0298614501953125,
0.023468017578125,
-0.0196990966796875,
-0.0004744529724121094,
0.024383544921875,
0.01010894775390625,
-0.0440673828125,
0.01149749755859375,
-0.03680419921875,
0.037353515625,
0.059173583984375,
-0.008453369140625,
0.026702880859375,
-0.036590576171875,
0.03131103515625,
0.0188140869140625,
0.0308380126953125,
-0.01294708251953125,
-0.05108642578125,
-0.07110595703125,
-0.029693603515625,
0.02655029296875,
0.0323486328125,
-0.064453125,
0.028656005859375,
-0.013916015625,
-0.053192138671875,
-0.046600341796875,
-0.0098876953125,
0.032745361328125,
0.0394287109375,
0.041259765625,
-0.02496337890625,
-0.048675537109375,
-0.060211181640625,
0.0014429092407226562,
0.0032253265380859375,
0.0020351409912109375,
0.0183258056640625,
0.04791259765625,
-0.022491455078125,
0.06536865234375,
-0.0254669189453125,
0.0008840560913085938,
-0.01995849609375,
0.0160064697265625,
0.04437255859375,
0.04595947265625,
0.042724609375,
-0.058563232421875,
-0.050140380859375,
-0.01346588134765625,
-0.048736572265625,
0.007656097412109375,
-0.001827239990234375,
-0.012725830078125,
0.01495361328125,
0.01180267333984375,
-0.07000732421875,
0.040130615234375,
0.04412841796875,
-0.039306640625,
0.045745849609375,
-0.0059051513671875,
0.007770538330078125,
-0.106201171875,
0.01251220703125,
-0.0023441314697265625,
-0.01337432861328125,
-0.03955078125,
-0.000946044921875,
0.0091552734375,
-0.005458831787109375,
-0.0552978515625,
0.03857421875,
-0.03375244140625,
-0.0010471343994140625,
-0.0169677734375,
-0.018035888671875,
-0.005985260009765625,
0.054931640625,
0.006378173828125,
0.061676025390625,
0.039215087890625,
-0.03680419921875,
0.03704833984375,
0.01800537109375,
-0.0170135498046875,
0.017059326171875,
-0.049041748046875,
0.0078887939453125,
0.0077972412109375,
0.0167236328125,
-0.06365966796875,
-0.012969970703125,
0.03411865234375,
-0.044219970703125,
0.0245361328125,
-0.0287322998046875,
-0.039642333984375,
-0.04644775390625,
-0.01389312744140625,
0.031951904296875,
0.050262451171875,
-0.05816650390625,
0.05029296875,
0.005275726318359375,
0.015777587890625,
-0.06134033203125,
-0.044464111328125,
-0.009857177734375,
-0.0209808349609375,
-0.05804443359375,
0.02667236328125,
-0.0030879974365234375,
0.01390838623046875,
-0.0098114013671875,
-0.01232147216796875,
0.0035552978515625,
-0.00885009765625,
0.0268707275390625,
0.02374267578125,
-0.0187835693359375,
-0.0172882080078125,
-0.017547607421875,
-0.02008056640625,
0.00824737548828125,
-0.0220489501953125,
0.072998046875,
-0.024810791015625,
-0.023773193359375,
-0.045166015625,
0.007358551025390625,
0.043426513671875,
-0.0158233642578125,
0.07928466796875,
0.0811767578125,
-0.01470947265625,
0.003025054931640625,
-0.0430908203125,
-0.02032470703125,
-0.038116455078125,
0.0287322998046875,
-0.0145111083984375,
-0.046173095703125,
0.048004150390625,
0.01485443115234375,
-0.007640838623046875,
0.05145263671875,
0.05596923828125,
-0.006603240966796875,
0.07574462890625,
0.038177490234375,
0.01119232177734375,
0.04205322265625,
-0.059722900390625,
-0.0030670166015625,
-0.0675048828125,
-0.02239990234375,
-0.01531219482421875,
-0.0203857421875,
-0.04974365234375,
-0.039947509765625,
0.0211944580078125,
-0.004711151123046875,
-0.04815673828125,
0.055084228515625,
-0.0439453125,
0.029693603515625,
0.059112548828125,
0.032623291015625,
0.0023212432861328125,
-0.00849151611328125,
-0.02093505859375,
0.013092041015625,
-0.06463623046875,
-0.0270233154296875,
0.09552001953125,
0.0298004150390625,
0.046295166015625,
0.00505828857421875,
0.053955078125,
-0.00716400146484375,
0.028656005859375,
-0.0292816162109375,
0.03509521484375,
0.0036907196044921875,
-0.052581787109375,
-0.00539398193359375,
-0.052886962890625,
-0.06646728515625,
0.017913818359375,
-0.0135040283203125,
-0.05670166015625,
0.0283203125,
0.01446533203125,
-0.0304107666015625,
0.044281005859375,
-0.0693359375,
0.07818603515625,
-0.01145172119140625,
-0.03411865234375,
0.00667572021484375,
-0.059600830078125,
0.027862548828125,
0.00917816162109375,
-0.00580596923828125,
-0.0023479461669921875,
0.01505279541015625,
0.0660400390625,
-0.0364990234375,
0.06597900390625,
-0.01397705078125,
0.017242431640625,
0.0263671875,
-0.007785797119140625,
0.0428466796875,
-0.0014200210571289062,
0.0008940696716308594,
0.01104736328125,
0.0099334716796875,
-0.032928466796875,
-0.0265350341796875,
0.036224365234375,
-0.08526611328125,
-0.039398193359375,
-0.0413818359375,
-0.0474853515625,
0.004161834716796875,
0.01404571533203125,
0.047882080078125,
0.0242919921875,
0.015594482421875,
0.0272216796875,
0.0478515625,
-0.03021240234375,
0.05291748046875,
0.0164642333984375,
-0.0067291259765625,
-0.0430908203125,
0.06927490234375,
0.0015087127685546875,
0.03338623046875,
0.016265869140625,
0.018524169921875,
-0.02655029296875,
-0.03277587890625,
-0.03424072265625,
0.0283203125,
-0.046661376953125,
-0.0277862548828125,
-0.053955078125,
-0.034820556640625,
-0.038055419921875,
0.00975799560546875,
-0.0419921875,
-0.0277862548828125,
-0.03265380859375,
-0.0020961761474609375,
0.0285797119140625,
0.035552978515625,
-0.005702972412109375,
0.048004150390625,
-0.054046630859375,
0.0228118896484375,
0.0188446044921875,
0.0303192138671875,
0.002140045166015625,
-0.0574951171875,
-0.0230712890625,
0.0161590576171875,
-0.043243408203125,
-0.0745849609375,
0.040863037109375,
-0.00316619873046875,
0.0374755859375,
0.021575927734375,
-0.007965087890625,
0.0477294921875,
-0.0173187255859375,
0.060760498046875,
0.0265960693359375,
-0.06329345703125,
0.0223846435546875,
-0.031829833984375,
0.033203125,
0.0124359130859375,
0.044464111328125,
-0.033966064453125,
-0.01788330078125,
-0.06597900390625,
-0.062347412109375,
0.07098388671875,
0.041107177734375,
0.0029659271240234375,
0.000888824462890625,
0.0220947265625,
-0.00457763671875,
0.00954437255859375,
-0.0924072265625,
-0.02032470703125,
-0.038330078125,
-0.02154541015625,
-0.0012216567993164062,
-0.0101318359375,
-0.005336761474609375,
-0.0433349609375,
0.0611572265625,
-0.0004143714904785156,
0.04876708984375,
0.015380859375,
-0.0190277099609375,
-0.0028438568115234375,
-0.0032787322998046875,
0.035614013671875,
0.0545654296875,
-0.0250701904296875,
0.004291534423828125,
0.01678466796875,
-0.05474853515625,
0.0025615692138671875,
0.0180511474609375,
-0.007568359375,
-0.012664794921875,
0.02227783203125,
0.06927490234375,
0.0003674030303955078,
-0.0175018310546875,
0.046722412109375,
-0.00261688232421875,
-0.013092041015625,
-0.01195526123046875,
0.00913238525390625,
0.02911376953125,
0.035675048828125,
0.018218994140625,
0.0026493072509765625,
-0.01279449462890625,
-0.0361328125,
0.020294189453125,
0.01528167724609375,
-0.01294708251953125,
-0.021453857421875,
0.0631103515625,
-0.00057220458984375,
-0.01467132568359375,
0.055450439453125,
-0.008087158203125,
-0.0386962890625,
0.0631103515625,
0.052459716796875,
0.059814453125,
-0.02362060546875,
0.0178070068359375,
0.03826904296875,
0.024932861328125,
-0.00225067138671875,
0.0113677978515625,
-0.00005030632019042969,
-0.04541015625,
-0.0292816162109375,
-0.060546875,
-0.016754150390625,
0.004924774169921875,
-0.03521728515625,
0.0322265625,
-0.032318115234375,
-0.0195465087890625,
-0.0094451904296875,
0.003963470458984375,
-0.051788330078125,
0.019866943359375,
0.01837158203125,
0.06988525390625,
-0.058746337890625,
0.06982421875,
0.029083251953125,
-0.0517578125,
-0.0791015625,
-0.0193634033203125,
-0.007015228271484375,
-0.06793212890625,
0.03045654296875,
0.02093505859375,
0.01436614990234375,
0.01050567626953125,
-0.048553466796875,
-0.07073974609375,
0.10986328125,
0.0416259765625,
-0.0234222412109375,
-0.020294189453125,
0.0322265625,
0.03607177734375,
-0.0252227783203125,
0.049285888671875,
0.037628173828125,
0.0273895263671875,
0.03082275390625,
-0.0631103515625,
0.0091400146484375,
-0.026275634765625,
-0.004878997802734375,
0.00684356689453125,
-0.061859130859375,
0.0904541015625,
-0.0116729736328125,
-0.00737762451171875,
0.008209228515625,
0.05010986328125,
0.0233612060546875,
0.01239013671875,
0.0251922607421875,
0.061126708984375,
0.031707763671875,
-0.019287109375,
0.08807373046875,
-0.0272216796875,
0.0511474609375,
0.07196044921875,
0.023834228515625,
0.043243408203125,
0.0285491943359375,
-0.0176849365234375,
0.031768798828125,
0.0682373046875,
-0.02532958984375,
0.029510498046875,
-0.0026187896728515625,
-0.007099151611328125,
-0.015594482421875,
0.0190277099609375,
-0.04437255859375,
0.027984619140625,
0.0102081298828125,
-0.048858642578125,
-0.01299285888671875,
0.006168365478515625,
0.0124969482421875,
-0.033905029296875,
-0.01038360595703125,
0.0447998046875,
0.009613037109375,
-0.037017822265625,
0.058563232421875,
-0.00341796875,
0.050048828125,
-0.039093017578125,
0.006320953369140625,
-0.0223236083984375,
0.0170135498046875,
-0.0221405029296875,
-0.057403564453125,
0.0183258056640625,
-0.0094757080078125,
-0.006206512451171875,
-0.0101318359375,
0.0238800048828125,
-0.0264892578125,
-0.035552978515625,
0.0185546875,
0.01922607421875,
0.0119171142578125,
-0.01204681396484375,
-0.0670166015625,
-0.007503509521484375,
0.0018529891967773438,
-0.032867431640625,
0.019287109375,
0.0165863037109375,
0.0220184326171875,
0.05108642578125,
0.0546875,
-0.005802154541015625,
0.0248565673828125,
-0.01053619384765625,
0.07525634765625,
-0.05169677734375,
-0.0286407470703125,
-0.0714111328125,
0.049591064453125,
-0.0045166015625,
-0.026397705078125,
0.0626220703125,
0.0537109375,
0.067626953125,
-0.0035533905029296875,
0.0271148681640625,
-0.0164642333984375,
0.0164337158203125,
-0.033111572265625,
0.065673828125,
-0.031890869140625,
0.01409149169921875,
-0.029022216796875,
-0.0863037109375,
-0.0108184814453125,
0.04168701171875,
-0.02752685546875,
0.0179901123046875,
0.05291748046875,
0.0670166015625,
-0.0252532958984375,
0.00621795654296875,
0.01180267333984375,
0.0296173095703125,
0.021209716796875,
0.056121826171875,
0.05755615234375,
-0.050933837890625,
0.052001953125,
-0.04754638671875,
-0.01213836669921875,
-0.00505828857421875,
-0.0546875,
-0.0771484375,
-0.04168701171875,
-0.0177154541015625,
-0.0445556640625,
-0.0022869110107421875,
0.07562255859375,
0.0631103515625,
-0.0517578125,
-0.025665283203125,
-0.0033111572265625,
0.00027680397033691406,
-0.01186370849609375,
-0.01617431640625,
0.042877197265625,
-0.00140380859375,
-0.052886962890625,
0.0020236968994140625,
0.002483367919921875,
0.0220947265625,
-0.0011663436889648438,
-0.00948333740234375,
-0.03729248046875,
-0.0008940696716308594,
0.035614013671875,
0.0171966552734375,
-0.041259765625,
-0.0181732177734375,
0.0054931640625,
-0.007144927978515625,
0.033233642578125,
0.0274658203125,
-0.045074462890625,
0.02142333984375,
0.025634765625,
0.030242919921875,
0.07501220703125,
-0.00437164306640625,
0.034637451171875,
-0.04180908203125,
0.0106048583984375,
0.017364501953125,
0.0384521484375,
0.0217742919921875,
-0.0269775390625,
0.041290283203125,
0.039093017578125,
-0.0380859375,
-0.055572509765625,
-0.0002875328063964844,
-0.08135986328125,
-0.00817108154296875,
0.0902099609375,
-0.0147247314453125,
-0.0421142578125,
0.0210113525390625,
-0.016754150390625,
0.048828125,
-0.01065826416015625,
0.050445556640625,
0.038330078125,
-0.0085906982421875,
-0.03839111328125,
-0.0244598388671875,
0.03448486328125,
0.0205535888671875,
-0.05291748046875,
-0.0123443603515625,
0.00917816162109375,
0.040679931640625,
0.01505279541015625,
0.02667236328125,
-0.01409149169921875,
0.034271240234375,
0.006000518798828125,
0.0178375244140625,
-0.0242919921875,
-0.0170135498046875,
-0.006481170654296875,
0.0108795166015625,
-0.0251312255859375,
-0.01220703125
]
] |
PygmalionAI/pygmalion-350m | 2023-01-11T23:44:13.000Z | [
"transformers",
"pytorch",
"opt",
"text-generation",
"convAI",
"conversational",
"en",
"has_space",
"text-generation-inference",
"region:us"
] | conversational | PygmalionAI | null | null | PygmalionAI/pygmalion-350m | 49 | 7,576 | transformers | 2022-12-20T22:04:32 | ---
language:
- en
thumbnail:
tags:
- convAI
- conversational
inference: false
---
# pygmalion-350m
# Model description
This is a proof-of-concept fine-tune of Facebook's OPT-350M model optimized for dialogue, to be used as a stepping stone to higher parameter models.
**Disclaimer:** NSFW data was included in the fine-tuning of this model. Although SFW inputs will usually result in SFW outputs, you are advised to **chat at your own risk. This model is not suitable for use by minors.**
# Fine-tuning process
This model was much easier than expected to create.
We used the [ColossalAI](https://www.colossalai.org/) library to fine-tune the [OPT-350M](https://huggingface.co/facebook/opt-350m) model originally trained by Facebook on The Pile. Though our initial dataset was sets of dialogue gathered from various sources totaling about 50 MB in size, early training runs revealed that the model converged after only 7% of the dataset was passed through. To alleviate this, we massively reduced the size of the dataset to only 273 KB.
ColossalAI's magic allowed for something incredible: this entire model was fine-tuned on a singular GPU with only 6 GB ***(!)*** of VRAM. Fine-tuning took less than an hour to complete. | 1,229 | [
[
-0.051666259765625,
-0.072998046875,
0.030029296875,
0.01605224609375,
-0.039154052734375,
-0.0184478759765625,
-0.029205322265625,
-0.0309295654296875,
0.01419830322265625,
0.0345458984375,
-0.0662841796875,
-0.006969451904296875,
-0.0267181396484375,
-0.01116180419921875,
-0.01114654541015625,
0.0869140625,
0.0121307373046875,
-0.00493621826171875,
0.011962890625,
0.0050811767578125,
-0.01617431640625,
-0.01198577880859375,
-0.0697021484375,
-0.01168060302734375,
0.05078125,
0.0189361572265625,
0.0804443359375,
0.0654296875,
0.0264434814453125,
0.0216217041015625,
-0.027496337890625,
0.0281219482421875,
-0.08349609375,
-0.0626220703125,
-0.0131378173828125,
-0.00988006591796875,
-0.029998779296875,
0.0194091796875,
0.059295654296875,
0.0545654296875,
-0.00910186767578125,
0.034149169921875,
-0.004364013671875,
0.034027099609375,
-0.03564453125,
0.0252838134765625,
-0.036102294921875,
0.000024199485778808594,
0.00563812255859375,
0.000606536865234375,
-0.0272216796875,
-0.023834228515625,
0.011260986328125,
-0.06109619140625,
0.03253173828125,
0.01267242431640625,
0.08709716796875,
0.016204833984375,
-0.028076171875,
0.00749969482421875,
-0.043853759765625,
0.0450439453125,
-0.059661865234375,
0.005218505859375,
0.0184478759765625,
0.020477294921875,
-0.021514892578125,
-0.046905517578125,
-0.031829833984375,
-0.0288543701171875,
0.01611328125,
-0.005279541015625,
-0.0167083740234375,
0.0301971435546875,
0.052947998046875,
0.049041748046875,
-0.04443359375,
-0.005950927734375,
-0.052642822265625,
-0.0160980224609375,
0.065673828125,
0.0211181640625,
0.01416778564453125,
-0.0207366943359375,
-0.0135040283203125,
-0.00482177734375,
-0.022613525390625,
0.010101318359375,
0.031158447265625,
0.0243682861328125,
-0.01092529296875,
0.040313720703125,
-0.04510498046875,
0.062164306640625,
0.026458740234375,
-0.01308441162109375,
0.0024242401123046875,
-0.0210418701171875,
-0.0254364013671875,
0.002750396728515625,
0.075439453125,
0.0523681640625,
0.0231170654296875,
-0.0014276504516601562,
-0.005970001220703125,
0.01690673828125,
0.037872314453125,
-0.08660888671875,
-0.03759765625,
0.007904052734375,
-0.0430908203125,
-0.01715087890625,
-0.006305694580078125,
-0.01708984375,
-0.0181427001953125,
-0.02099609375,
0.02728271484375,
-0.031890869140625,
-0.0355224609375,
0.0208282470703125,
-0.0022792816162109375,
-0.0014324188232421875,
0.03082275390625,
-0.063232421875,
0.009490966796875,
0.045318603515625,
0.06805419921875,
0.00688934326171875,
-0.00962066650390625,
-0.0240478515625,
0.0027904510498046875,
-0.03558349609375,
0.0323486328125,
-0.0171661376953125,
-0.033905029296875,
-0.0020751953125,
0.01244354248046875,
-0.0018587112426757812,
-0.039947509765625,
0.057708740234375,
-0.024932861328125,
-0.00872802734375,
-0.0099029541015625,
-0.03765869140625,
-0.041412353515625,
-0.0014781951904296875,
-0.057464599609375,
0.06048583984375,
0.0218353271484375,
-0.039337158203125,
0.033599853515625,
-0.0294342041015625,
-0.0191802978515625,
0.021514892578125,
0.003391265869140625,
-0.0178985595703125,
0.031463623046875,
0.021728515625,
0.043853759765625,
-0.0347900390625,
0.03173828125,
-0.048065185546875,
-0.04827880859375,
0.021697998046875,
-0.02777099609375,
0.03875732421875,
0.0248870849609375,
-0.004634857177734375,
0.01293182373046875,
-0.0472412109375,
0.00347137451171875,
0.0032062530517578125,
-0.027252197265625,
0.025360107421875,
-0.00983428955078125,
0.0169677734375,
0.0245208740234375,
0.031402587890625,
-0.045166015625,
0.00379180908203125,
-0.032958984375,
0.0411376953125,
0.049530029296875,
-0.005229949951171875,
0.0214080810546875,
-0.04022216796875,
0.022735595703125,
-0.01556396484375,
0.039825439453125,
-0.001056671142578125,
-0.04583740234375,
-0.04022216796875,
-0.0291595458984375,
-0.0102386474609375,
0.01171112060546875,
-0.045501708984375,
0.00753021240234375,
-0.015777587890625,
-0.04754638671875,
-0.02728271484375,
0.0179901123046875,
0.03411865234375,
0.01666259765625,
-0.007434844970703125,
-0.0218963623046875,
-0.0291900634765625,
-0.08319091796875,
-0.0169830322265625,
-0.0224456787109375,
-0.00545501708984375,
0.05450439453125,
0.0208282470703125,
-0.0243682861328125,
0.054107666015625,
-0.0574951171875,
0.0002696514129638672,
-0.011749267578125,
0.002574920654296875,
0.042083740234375,
0.046875,
0.03839111328125,
-0.059051513671875,
-0.0272674560546875,
0.00411224365234375,
-0.049041748046875,
0.003108978271484375,
-0.004180908203125,
-0.0158843994140625,
-0.005260467529296875,
0.02569580078125,
-0.06427001953125,
0.0295867919921875,
0.035980224609375,
-0.0117645263671875,
0.036346435546875,
-0.0197906494140625,
0.0101318359375,
-0.0802001953125,
-0.011260986328125,
0.031402587890625,
-0.022186279296875,
-0.0194854736328125,
-0.0013065338134765625,
0.00775146484375,
-0.03533935546875,
-0.0379638671875,
0.01485443115234375,
-0.0260772705078125,
0.0248565673828125,
-0.026580810546875,
-0.002544403076171875,
-0.0283050537109375,
0.060821533203125,
-0.0063323974609375,
0.04681396484375,
0.044708251953125,
-0.055267333984375,
0.039215087890625,
0.017547607421875,
-0.0189666748046875,
0.032470703125,
-0.057952880859375,
0.007160186767578125,
0.002788543701171875,
0.0144195556640625,
-0.0811767578125,
-0.05474853515625,
0.0197601318359375,
-0.049591064453125,
0.009429931640625,
-0.01224517822265625,
-0.037139892578125,
-0.043212890625,
-0.018646240234375,
0.033782958984375,
0.03472900390625,
-0.059234619140625,
0.045928955078125,
0.03466796875,
-0.003894805908203125,
-0.0012445449829101562,
-0.0257110595703125,
-0.000560760498046875,
-0.019317626953125,
-0.0755615234375,
0.014556884765625,
-0.01305389404296875,
0.01214599609375,
-0.00725555419921875,
-0.00833892822265625,
-0.00553131103515625,
-0.0216064453125,
0.0472412109375,
0.027587890625,
-0.0178985595703125,
-0.017059326171875,
0.0081939697265625,
-0.004150390625,
0.007720947265625,
-0.0166015625,
0.03265380859375,
-0.0270233154296875,
0.004673004150390625,
-0.0697021484375,
0.0020580291748046875,
0.04046630859375,
0.0141143798828125,
0.05517578125,
0.05364990234375,
-0.04052734375,
-0.01293182373046875,
-0.054046630859375,
-0.0071258544921875,
-0.042938232421875,
0.03692626953125,
-0.0204315185546875,
-0.07635498046875,
0.03228759765625,
-0.03472900390625,
-0.00634765625,
0.0345458984375,
0.03765869140625,
0.01049041748046875,
0.11175537109375,
0.046905517578125,
-0.01605224609375,
0.048736572265625,
-0.0211181640625,
0.006458282470703125,
-0.05767822265625,
-0.01263427734375,
-0.0140228271484375,
-0.00994873046875,
-0.05450439453125,
-0.0309600830078125,
0.020050048828125,
0.00713348388671875,
-0.030670166015625,
0.043975830078125,
-0.0404052734375,
0.023529052734375,
0.028076171875,
0.0328369140625,
-0.00914764404296875,
0.00815582275390625,
0.0179443359375,
0.003726959228515625,
-0.0618896484375,
-0.035400390625,
0.0703125,
0.0290069580078125,
0.05706787109375,
-0.00919342041015625,
0.0194854736328125,
0.0095367431640625,
0.022552490234375,
-0.053131103515625,
0.064208984375,
-0.006198883056640625,
-0.06707763671875,
-0.0177764892578125,
-0.057373046875,
-0.047454833984375,
0.0201568603515625,
-0.007274627685546875,
-0.06854248046875,
-0.006500244140625,
0.013153076171875,
-0.039642333984375,
0.0146026611328125,
-0.07318115234375,
0.07720947265625,
0.005832672119140625,
-0.03515625,
-0.0269927978515625,
-0.03167724609375,
0.03564453125,
0.003932952880859375,
0.00487518310546875,
-0.025177001953125,
0.0185546875,
0.0594482421875,
-0.043304443359375,
0.052734375,
-0.00885772705078125,
-0.004486083984375,
0.0287322998046875,
-0.0023136138916015625,
0.0282440185546875,
0.031494140625,
0.0140228271484375,
0.0015583038330078125,
-0.0011749267578125,
-0.036102294921875,
-0.019073486328125,
0.0626220703125,
-0.07281494140625,
-0.034332275390625,
-0.015106201171875,
-0.054840087890625,
-0.0235748291015625,
0.00452423095703125,
0.0220794677734375,
0.033721923828125,
-0.0207061767578125,
0.0290069580078125,
0.053314208984375,
-0.0133819580078125,
0.036224365234375,
0.034027099609375,
-0.0169830322265625,
-0.049591064453125,
0.0787353515625,
0.004486083984375,
0.0178375244140625,
-0.0094757080078125,
0.01666259765625,
-0.0204620361328125,
-0.0230255126953125,
-0.045379638671875,
0.0333251953125,
-0.0289306640625,
0.0006604194641113281,
-0.034515380859375,
-0.014404296875,
-0.033416748046875,
-0.016448974609375,
-0.036285400390625,
-0.00896453857421875,
-0.034912109375,
-0.005825042724609375,
0.0276641845703125,
0.043975830078125,
-0.0005559921264648438,
0.051483154296875,
-0.041595458984375,
0.04608154296875,
0.0203857421875,
0.03729248046875,
-0.0100555419921875,
-0.054656982421875,
-0.0277099609375,
0.0271148681640625,
-0.0162353515625,
-0.04052734375,
0.032928466796875,
0.01503753662109375,
0.0296478271484375,
0.05694580078125,
0.001827239990234375,
0.069580078125,
-0.018890380859375,
0.05804443359375,
0.0294189453125,
-0.053192138671875,
0.031707763671875,
-0.05987548828125,
0.0292816162109375,
0.033294677734375,
0.0242919921875,
-0.02435302734375,
-0.01580810546875,
-0.058013916015625,
-0.059539794921875,
0.0924072265625,
0.013763427734375,
0.020965576171875,
0.01155853271484375,
0.046142578125,
-0.00774383544921875,
0.017486572265625,
-0.053314208984375,
0.0044708251953125,
-0.0142974853515625,
-0.00907135009765625,
-0.0306549072265625,
-0.0260772705078125,
-0.018798828125,
-0.0460205078125,
0.0645751953125,
-0.023712158203125,
0.0022792816162109375,
-0.0031604766845703125,
-0.00803375244140625,
-0.0293121337890625,
-0.0352783203125,
0.03375244140625,
0.061309814453125,
-0.028839111328125,
-0.032501220703125,
0.0131988525390625,
-0.045013427734375,
-0.023834228515625,
0.017578125,
-0.0013294219970703125,
-0.0082244873046875,
-0.0025081634521484375,
0.09765625,
0.017425537109375,
-0.052581787109375,
0.024658203125,
-0.01233673095703125,
-0.01629638671875,
-0.019989013671875,
0.0270843505859375,
0.017669677734375,
0.031646728515625,
0.027862548828125,
-0.007129669189453125,
0.017791748046875,
-0.032958984375,
-0.0027618408203125,
0.025634765625,
-0.03057861328125,
-0.00885009765625,
0.06298828125,
-0.00469970703125,
-0.020416259765625,
0.052581787109375,
-0.011749267578125,
-0.00885009765625,
0.048248291015625,
0.031890869140625,
0.057281494140625,
-0.00585174560546875,
0.029693603515625,
0.048248291015625,
0.0261993408203125,
-0.0214996337890625,
0.01507568359375,
-0.007762908935546875,
-0.03533935546875,
-0.0311279296875,
-0.06402587890625,
-0.0286865234375,
0.0195465087890625,
-0.06048583984375,
0.006755828857421875,
-0.0631103515625,
-0.024078369140625,
0.0244293212890625,
0.007411956787109375,
-0.0285797119140625,
0.0570068359375,
0.0131988525390625,
0.06024169921875,
-0.0570068359375,
0.04217529296875,
0.056396484375,
-0.035858154296875,
-0.08892822265625,
-0.0123138427734375,
-0.006122589111328125,
-0.047027587890625,
0.021270751953125,
0.0394287109375,
0.004589080810546875,
0.0034542083740234375,
-0.08251953125,
-0.05096435546875,
0.06817626953125,
0.050262451171875,
-0.04730224609375,
0.00200653076171875,
0.0177764892578125,
0.03564453125,
-0.020263671875,
0.01023101806640625,
0.0333251953125,
0.00925445556640625,
0.0297698974609375,
-0.08843994140625,
0.0150604248046875,
-0.0325927734375,
0.033416748046875,
-0.002094268798828125,
-0.0748291015625,
0.08251953125,
0.000274658203125,
0.002288818359375,
0.0198516845703125,
0.047607421875,
0.01349639892578125,
0.01541900634765625,
0.039794921875,
0.05072021484375,
0.035797119140625,
-0.0139617919921875,
0.081298828125,
-0.026397705078125,
0.0382080078125,
0.0474853515625,
-0.0013103485107421875,
0.0308685302734375,
0.0121917724609375,
-0.0030956268310546875,
-0.004734039306640625,
0.07818603515625,
0.0013742446899414062,
0.0201568603515625,
0.0267791748046875,
-0.0265655517578125,
-0.04052734375,
-0.0165557861328125,
-0.024261474609375,
0.01221466064453125,
0.0292816162109375,
-0.0235137939453125,
0.0038394927978515625,
0.0098419189453125,
-0.00040459632873535156,
-0.019500732421875,
-0.05517578125,
0.06488037109375,
-0.0005035400390625,
-0.05816650390625,
0.068359375,
0.00366973876953125,
0.046844482421875,
-0.041168212890625,
0.01258087158203125,
-0.01027679443359375,
0.0333251953125,
-0.0150299072265625,
-0.0340576171875,
-0.00251007080078125,
-0.01297760009765625,
-0.00661468505859375,
-0.003910064697265625,
0.051605224609375,
-0.0169525146484375,
-0.046722412109375,
0.01049041748046875,
-0.0012874603271484375,
0.02056884765625,
-0.046539306640625,
-0.0556640625,
0.020172119140625,
-0.0017147064208984375,
-0.01910400390625,
0.01096343994140625,
0.03662109375,
-0.002521514892578125,
0.059173583984375,
0.060638427734375,
-0.008636474609375,
0.00434112548828125,
-0.006717681884765625,
0.08978271484375,
-0.042449951171875,
-0.0526123046875,
-0.050140380859375,
0.0190887451171875,
-0.017333984375,
-0.04296875,
0.06719970703125,
0.0560302734375,
0.07135009765625,
-0.02020263671875,
0.038360595703125,
-0.004734039306640625,
0.019287109375,
-0.0195465087890625,
0.048980712890625,
-0.055267333984375,
-0.0269775390625,
-0.0469970703125,
-0.056549072265625,
0.0115814208984375,
0.06561279296875,
-0.0063629150390625,
0.01256561279296875,
0.03936767578125,
0.078857421875,
-0.0024623870849609375,
0.037139892578125,
0.02813720703125,
0.01276397705078125,
0.007640838623046875,
0.0499267578125,
0.0645751953125,
-0.0248870849609375,
0.050811767578125,
-0.026397705078125,
-0.035675048828125,
-0.0268707275390625,
-0.044769287109375,
-0.080322265625,
-0.0352783203125,
-0.040740966796875,
-0.04083251953125,
0.00214385986328125,
0.04766845703125,
0.0784912109375,
-0.03790283203125,
-0.029144287109375,
-0.0132293701171875,
-0.0283203125,
-0.00972747802734375,
-0.01349639892578125,
0.001468658447265625,
0.002410888671875,
-0.05035400390625,
0.0169219970703125,
0.0147552490234375,
-0.0187530517578125,
-0.0143890380859375,
-0.00588226318359375,
-0.0165557861328125,
0.01381683349609375,
0.032135009765625,
0.020111083984375,
-0.041656494140625,
-0.0192108154296875,
-0.0164794921875,
-0.00534820556640625,
-0.0030803680419921875,
0.055267333984375,
-0.035888671875,
0.0288543701171875,
0.0234222412109375,
0.038787841796875,
0.06854248046875,
0.004367828369140625,
0.0443115234375,
-0.03021240234375,
0.026397705078125,
0.0011682510375976562,
0.0290985107421875,
0.0175323486328125,
-0.044219970703125,
0.025970458984375,
0.033050537109375,
-0.039398193359375,
-0.06195068359375,
0.008209228515625,
-0.09539794921875,
-0.0166473388671875,
0.09454345703125,
-0.0009737014770507812,
-0.0293121337890625,
0.0197906494140625,
-0.0693359375,
0.0377197265625,
-0.0357666015625,
0.057861328125,
0.053558349609375,
0.00254058837890625,
-0.01297760009765625,
-0.050994873046875,
0.037384033203125,
0.0257568359375,
-0.0306549072265625,
-0.014404296875,
0.06805419921875,
0.0222625732421875,
0.01003265380859375,
0.037109375,
-0.00048065185546875,
0.042144775390625,
0.0135345458984375,
-0.016571044921875,
-0.0100250244140625,
-0.0111541748046875,
-0.00606536865234375,
0.03179931640625,
-0.01300048828125,
-0.0219879150390625
]
] |
ogkalu/Comic-Diffusion | 2023-05-10T17:20:27.000Z | [
"diffusers",
"text-to-image",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | ogkalu | null | null | ogkalu/Comic-Diffusion | 466 | 7,571 | diffusers | 2022-10-28T15:27:32 | ---
license: creativeml-openrail-m
tags:
- text-to-image
---
V2 is here. Trained on 6 styles at once, it allows anyone to create unique but consistent styles by mixing any number of the tokens. Even changing the order of the same list influences results so there's a lot to experiment with here. This was created so anyone could create their comic projects with ease and flexibility. It is the culmination of all my experimentation with dreambooth thus far.
The tokens for V2 are -
- charliebo artstyle
- holliemengert artstyle
- marioalberti artstyle
- pepelarraz artstyle
- andreasrocha artstyle
- jamesdaly artstyle
None of the artists used are affiliated with this.
**Generated by V2:**
V1 was trained solely on James Daly 3. He is **not** affiliated with this. The correct token for V1 is comicmay artsyle.
**Generated by V1:**
 | 1,597 | [
[
-0.0166473388671875,
-0.0278472900390625,
0.0182342529296875,
0.0198822021484375,
-0.03662109375,
0.0275726318359375,
0.01515960693359375,
-0.046417236328125,
0.0870361328125,
0.036376953125,
-0.05963134765625,
-0.017059326171875,
-0.048675537109375,
0.003803253173828125,
-0.01551055908203125,
0.057861328125,
-0.00823211669921875,
0.0004627704620361328,
0.00616455078125,
-0.0180206298828125,
-0.04742431640625,
-0.00476837158203125,
-0.0484619140625,
-0.0206298828125,
0.030792236328125,
0.03912353515625,
0.0531005859375,
0.0308685302734375,
0.0247344970703125,
0.01666259765625,
-0.00042891502380371094,
-0.007740020751953125,
-0.049163818359375,
0.01093292236328125,
-0.01311492919921875,
-0.0511474609375,
-0.043304443359375,
0.01224517822265625,
0.0341796875,
0.0187530517578125,
0.0013132095336914062,
0.017333984375,
-0.01812744140625,
0.054229736328125,
-0.02880859375,
-0.0029296875,
-0.0374755859375,
0.026580810546875,
-0.0279541015625,
0.0031642913818359375,
-0.0165863037109375,
-0.015380859375,
-0.031768798828125,
-0.055877685546875,
0.0234222412109375,
0.01427459716796875,
0.07037353515625,
0.0067138671875,
-0.019073486328125,
0.00714874267578125,
-0.033233642578125,
0.0256500244140625,
-0.020355224609375,
0.0220489501953125,
0.01513671875,
0.036712646484375,
-0.00859832763671875,
-0.09112548828125,
-0.06536865234375,
0.034027099609375,
0.01025390625,
0.037261962890625,
-0.0214691162109375,
-0.006908416748046875,
0.0171661376953125,
0.0238189697265625,
-0.033355712890625,
0.00482940673828125,
-0.059783935546875,
-0.0174102783203125,
0.036163330078125,
0.01361083984375,
0.0131378173828125,
-0.027313232421875,
-0.052459716796875,
-0.0228118896484375,
-0.051177978515625,
0.01506805419921875,
0.0498046875,
0.006580352783203125,
-0.0217742919921875,
0.055084228515625,
0.0053253173828125,
0.011962890625,
0.021392822265625,
-0.0201416015625,
0.051300048828125,
-0.03594970703125,
-0.0068359375,
-0.006298065185546875,
0.06591796875,
0.05230712890625,
0.005443572998046875,
0.004383087158203125,
-0.00923919677734375,
0.0001817941665649414,
0.038055419921875,
-0.08056640625,
-0.0211944580078125,
0.026641845703125,
-0.042724609375,
-0.046600341796875,
0.0009870529174804688,
-0.04449462890625,
-0.051666259765625,
-0.019683837890625,
0.0277099609375,
-0.036407470703125,
-0.060760498046875,
0.014617919921875,
-0.046630859375,
0.0166473388671875,
0.0218048095703125,
-0.04852294921875,
0.01078033447265625,
0.0399169921875,
0.05889892578125,
-0.0070343017578125,
-0.00037860870361328125,
-0.016143798828125,
-0.0167694091796875,
-0.054901123046875,
0.034820556640625,
-0.02569580078125,
-0.005367279052734375,
0.00588226318359375,
0.0207366943359375,
0.0144805908203125,
-0.043975830078125,
0.0214996337890625,
-0.02410888671875,
-0.006214141845703125,
-0.0223846435546875,
-0.028594970703125,
-0.0084991455078125,
-0.0175323486328125,
-0.051544189453125,
0.06378173828125,
0.0330810546875,
-0.0582275390625,
0.047271728515625,
-0.0299530029296875,
-0.0186767578125,
0.019134521484375,
-0.015411376953125,
-0.0255279541015625,
0.007289886474609375,
-0.031524658203125,
0.025970458984375,
0.0072784423828125,
-0.010345458984375,
-0.041046142578125,
-0.04119873046875,
-0.01406097412109375,
0.0100860595703125,
0.07562255859375,
0.0190277099609375,
-0.041259765625,
0.0091094970703125,
-0.061553955078125,
0.003208160400390625,
0.048736572265625,
0.017486572265625,
-0.007022857666015625,
-0.02618408203125,
0.0312347412109375,
0.032623291015625,
0.00841522216796875,
-0.040069580078125,
0.034027099609375,
-0.01404571533203125,
0.008544921875,
0.065673828125,
0.005817413330078125,
0.0041046142578125,
-0.036102294921875,
0.06298828125,
0.0100250244140625,
0.00891876220703125,
0.01404571533203125,
-0.0633544921875,
-0.06256103515625,
-0.0092010498046875,
-0.0108795166015625,
0.03668212890625,
-0.06475830078125,
0.0179901123046875,
0.007732391357421875,
-0.051788330078125,
-0.051239013671875,
0.00168609619140625,
0.0260162353515625,
-0.002300262451171875,
0.0150299072265625,
-0.057830810546875,
-0.053924560546875,
-0.06402587890625,
-0.00984954833984375,
-0.024169921875,
0.01389312744140625,
-0.0119171142578125,
0.0282440185546875,
-0.01514434814453125,
0.06658935546875,
-0.01322174072265625,
0.002101898193359375,
-0.0260162353515625,
0.004730224609375,
0.037628173828125,
0.0343017578125,
0.0787353515625,
-0.04541015625,
-0.026519775390625,
-0.0222015380859375,
-0.03314208984375,
-0.0092926025390625,
-0.004726409912109375,
-0.01849365234375,
-0.00003135204315185547,
0.0253143310546875,
-0.038238525390625,
0.0450439453125,
0.0238037109375,
-0.05157470703125,
0.0633544921875,
0.0026607513427734375,
0.01016998291015625,
-0.10577392578125,
-0.01117706298828125,
-0.003963470458984375,
-0.025726318359375,
-0.034088134765625,
0.0223236083984375,
-0.00565338134765625,
0.0045928955078125,
-0.042388916015625,
0.054107666015625,
-0.016998291015625,
0.005290985107421875,
-0.0212860107421875,
-0.004962921142578125,
0.006397247314453125,
0.02716064453125,
-0.006771087646484375,
0.0396728515625,
0.054046630859375,
-0.0164337158203125,
0.042572021484375,
0.01690673828125,
-0.024566650390625,
0.0516357421875,
-0.0582275390625,
0.038665771484375,
-0.0274200439453125,
0.021331787109375,
-0.10809326171875,
-0.015777587890625,
0.058074951171875,
-0.0262603759765625,
0.034423828125,
-0.0209503173828125,
-0.0601806640625,
-0.051483154296875,
-0.026336669921875,
0.033233642578125,
0.06597900390625,
-0.036895751953125,
0.01226806640625,
0.0158538818359375,
0.004909515380859375,
-0.050567626953125,
-0.040130615234375,
-0.0025234222412109375,
-0.01446533203125,
-0.04290771484375,
0.0369873046875,
-0.0285491943359375,
-0.023101806640625,
-0.006732940673828125,
-0.03167724609375,
0.006824493408203125,
-0.019866943359375,
0.0399169921875,
0.006359100341796875,
-0.01220703125,
-0.01470184326171875,
0.0224151611328125,
0.0031604766845703125,
-0.0016260147094726562,
-0.0306549072265625,
0.046844482421875,
-0.0086822509765625,
-0.025665283203125,
-0.048736572265625,
0.0164031982421875,
0.043914794921875,
0.006893157958984375,
0.041900634765625,
0.087890625,
-0.03076171875,
0.00875091552734375,
-0.0450439453125,
-0.00458526611328125,
-0.039276123046875,
0.014007568359375,
-0.03802490234375,
-0.061614990234375,
0.03448486328125,
0.007205963134765625,
0.0209503173828125,
0.049285888671875,
0.044586181640625,
-0.01506805419921875,
0.048065185546875,
0.043548583984375,
0.01245880126953125,
0.04559326171875,
-0.042938232421875,
0.0111846923828125,
-0.029815673828125,
-0.0299835205078125,
-0.007537841796875,
-0.056427001953125,
-0.04718017578125,
-0.047210693359375,
0.00635528564453125,
0.03192138671875,
-0.0164642333984375,
0.0582275390625,
-0.033966064453125,
0.038177490234375,
0.0294036865234375,
0.01328277587890625,
-0.0037059783935546875,
0.0199127197265625,
-0.01947021484375,
-0.003475189208984375,
-0.028411865234375,
-0.00040411949157714844,
0.0784912109375,
0.0273895263671875,
0.041259765625,
0.01473236083984375,
0.071533203125,
0.00579071044921875,
-0.004058837890625,
-0.062042236328125,
0.0457763671875,
0.0077056884765625,
-0.045562744140625,
0.0012073516845703125,
0.00371551513671875,
-0.0731201171875,
0.015655517578125,
-0.023834228515625,
-0.0697021484375,
0.0247039794921875,
0.0041351318359375,
-0.0023059844970703125,
0.03204345703125,
-0.038177490234375,
0.047760009765625,
0.021759033203125,
-0.040069580078125,
-0.0301361083984375,
-0.07037353515625,
0.01502227783203125,
-0.004619598388671875,
0.0118255615234375,
-0.01605224609375,
0.006649017333984375,
0.0584716796875,
-0.04730224609375,
0.0623779296875,
-0.01023101806640625,
0.00811767578125,
0.042510986328125,
0.01267242431640625,
0.036285400390625,
0.012481689453125,
0.007030487060546875,
0.004840850830078125,
-0.0011386871337890625,
-0.0399169921875,
-0.0239715576171875,
0.029052734375,
-0.058563232421875,
-0.03546142578125,
-0.019744873046875,
-0.00801849365234375,
0.0031604766845703125,
0.0295257568359375,
0.05938720703125,
0.042755126953125,
0.01507568359375,
0.0212554931640625,
0.03179931640625,
0.01314544677734375,
0.035736083984375,
0.007152557373046875,
-0.036346435546875,
-0.0435791015625,
0.047760009765625,
-0.006069183349609375,
0.030792236328125,
-0.0003719329833984375,
0.0189056396484375,
-0.0330810546875,
-0.02801513671875,
-0.032135009765625,
0.033538818359375,
-0.059417724609375,
-0.00989532470703125,
-0.03936767578125,
-0.0123748779296875,
-0.05224609375,
-0.0159454345703125,
-0.04541015625,
-0.052215576171875,
-0.01349639892578125,
-0.01543426513671875,
0.05328369140625,
0.0758056640625,
-0.0212249755859375,
0.01471710205078125,
-0.0137786865234375,
0.03717041015625,
0.0211029052734375,
0.0244598388671875,
-0.0328369140625,
-0.035369873046875,
-0.0005950927734375,
-0.0240325927734375,
-0.0240325927734375,
-0.08258056640625,
0.01129913330078125,
-0.01534271240234375,
0.024078369140625,
0.043304443359375,
-0.01336669921875,
0.05224609375,
-0.033294677734375,
0.0675048828125,
0.062103271484375,
-0.032989501953125,
0.045989990234375,
-0.037384033203125,
-0.00559234619140625,
0.04962158203125,
0.026458740234375,
-0.05096435546875,
-0.037689208984375,
-0.07391357421875,
-0.06707763671875,
0.04742431640625,
0.0333251953125,
0.019989013671875,
0.001026153564453125,
0.03057861328125,
0.0253448486328125,
0.0243988037109375,
-0.04156494140625,
-0.06402587890625,
-0.048583984375,
0.00588226318359375,
0.0196380615234375,
-0.01275634765625,
-0.007244110107421875,
-0.037506103515625,
0.04510498046875,
-0.015350341796875,
0.031951904296875,
-0.0022449493408203125,
0.030548095703125,
-0.0214996337890625,
-0.0124053955078125,
0.0189056396484375,
0.060516357421875,
-0.014068603515625,
-0.010833740234375,
-0.0241851806640625,
-0.057342529296875,
0.010040283203125,
-0.010345458984375,
0.0084991455078125,
0.02655029296875,
0.00873565673828125,
0.078857421875,
-0.01337432861328125,
-0.02294921875,
0.05401611328125,
-0.0030364990234375,
-0.0274505615234375,
-0.059600830078125,
0.0240325927734375,
-0.004688262939453125,
0.042724609375,
0.00910186767578125,
0.03466796875,
0.00727081298828125,
-0.041900634765625,
0.018463134765625,
0.00795745849609375,
-0.037353515625,
-0.0255584716796875,
0.072021484375,
-0.0155181884765625,
-0.02655029296875,
0.031829833984375,
-0.047332763671875,
-0.007415771484375,
0.064697265625,
0.07025146484375,
0.070068359375,
0.0007481575012207031,
0.0254364013671875,
0.047088623046875,
0.0108489990234375,
0.0004203319549560547,
0.053192138671875,
0.022003173828125,
-0.0413818359375,
-0.01519012451171875,
-0.033233642578125,
-0.0272216796875,
0.0198822021484375,
-0.044769287109375,
0.0635986328125,
-0.04632568359375,
-0.0214385986328125,
-0.0103607177734375,
-0.0081787109375,
-0.07342529296875,
0.0117645263671875,
0.00859832763671875,
0.082275390625,
-0.05511474609375,
0.044830322265625,
0.0738525390625,
-0.0289764404296875,
-0.07427978515625,
-0.00128936767578125,
0.034759521484375,
-0.0802001953125,
0.030731201171875,
0.0255889892578125,
0.0077056884765625,
0.000232696533203125,
-0.0457763671875,
-0.052642822265625,
0.09100341796875,
0.01084136962890625,
-0.033203125,
0.007663726806640625,
-0.01430511474609375,
0.033294677734375,
-0.0192108154296875,
0.0275421142578125,
0.0330810546875,
0.015594482421875,
0.049896240234375,
-0.06475830078125,
0.00473785400390625,
-0.05413818359375,
0.006862640380859375,
-0.00010007619857788086,
-0.07183837890625,
0.045257568359375,
-0.04449462890625,
-0.01374053955078125,
0.0258941650390625,
0.060546875,
0.036163330078125,
0.01027679443359375,
0.056182861328125,
0.042510986328125,
0.056884765625,
-0.004810333251953125,
0.072509765625,
-0.027435302734375,
0.0290069580078125,
0.0513916015625,
0.0134124755859375,
0.0357666015625,
0.025421142578125,
-0.027252197265625,
0.060760498046875,
0.0697021484375,
0.005916595458984375,
0.03729248046875,
0.0309295654296875,
0.0010776519775390625,
-0.01190185546875,
-0.005947113037109375,
-0.052886962890625,
0.024749755859375,
0.01171112060546875,
-0.0117340087890625,
0.0181732177734375,
-0.004802703857421875,
0.0271148681640625,
-0.032958984375,
-0.0101165771484375,
0.0263214111328125,
0.003490447998046875,
-0.0154571533203125,
0.0523681640625,
-0.01023101806640625,
0.04742431640625,
-0.047943115234375,
-0.01374053955078125,
-0.0258026123046875,
-0.0164642333984375,
-0.016754150390625,
-0.06805419921875,
-0.0006856918334960938,
-0.0258636474609375,
0.00710296630859375,
-0.0304107666015625,
0.057891845703125,
-0.0107421875,
-0.037322998046875,
0.01107025146484375,
0.020599365234375,
0.0300445556640625,
0.0166778564453125,
-0.04071044921875,
0.0036468505859375,
-0.019439697265625,
-0.017364501953125,
0.0112762451171875,
0.0239410400390625,
0.027374267578125,
0.01532745361328125,
0.01395416259765625,
0.00927734375,
-0.0025005340576171875,
0.029693603515625,
0.04833984375,
-0.0280609130859375,
-0.03948974609375,
-0.038818359375,
0.038665771484375,
-0.014678955078125,
-0.046295166015625,
0.0249786376953125,
0.039031982421875,
0.0655517578125,
-0.02203369140625,
0.051605224609375,
-0.0269012451171875,
0.008087158203125,
-0.0343017578125,
0.051177978515625,
-0.08489990234375,
-0.0274810791015625,
-0.045196533203125,
-0.06414794921875,
0.006282806396484375,
0.032958984375,
0.030242919921875,
0.00197601318359375,
0.06011962890625,
0.043365478515625,
0.0029926300048828125,
0.004779815673828125,
0.0008606910705566406,
0.00386810302734375,
0.004131317138671875,
0.0153961181640625,
0.052398681640625,
-0.03472900390625,
0.0169525146484375,
-0.037689208984375,
-0.01251220703125,
-0.0218505859375,
-0.04815673828125,
-0.0709228515625,
-0.057281494140625,
-0.033172607421875,
-0.06982421875,
-0.0017900466918945312,
0.0850830078125,
0.06329345703125,
-0.042938232421875,
-0.0024852752685546875,
-0.0019664764404296875,
0.00830078125,
-0.035125732421875,
-0.0222015380859375,
0.00778961181640625,
0.00009840726852416992,
-0.06732177734375,
0.033294677734375,
0.026397705078125,
0.02178955078125,
-0.0033130645751953125,
0.01395416259765625,
0.0389404296875,
-0.01201629638671875,
0.04351806640625,
0.00966644287109375,
-0.0555419921875,
-0.04290771484375,
-0.005397796630859375,
-0.005924224853515625,
0.01337432861328125,
0.059600830078125,
-0.052032470703125,
0.048797607421875,
0.060577392578125,
0.018798828125,
0.0377197265625,
-0.0211334228515625,
0.00933074951171875,
-0.061553955078125,
0.01568603515625,
-0.003387451171875,
0.04632568359375,
-0.0032291412353515625,
-0.018341064453125,
0.056304931640625,
0.08013916015625,
-0.00919342041015625,
-0.03717041015625,
0.0271759033203125,
-0.0992431640625,
-0.00833892822265625,
0.09759521484375,
0.026123046875,
0.003971099853515625,
0.00019299983978271484,
-0.026580810546875,
0.0142364501953125,
-0.0160980224609375,
0.02508544921875,
0.040252685546875,
-0.0189666748046875,
-0.01165008544921875,
-0.05316162109375,
0.0232391357421875,
0.0095672607421875,
-0.034149169921875,
-0.031280517578125,
0.024078369140625,
0.03802490234375,
0.0240325927734375,
0.0799560546875,
-0.0094146728515625,
0.021820068359375,
0.024810791015625,
0.01006317138671875,
-0.00394439697265625,
-0.04266357421875,
0.011566162109375,
0.0099945068359375,
-0.00801849365234375,
-0.02008056640625
]
] |
ArthurZ/tiny-random-bert-sharded | 2022-06-17T08:07:42.000Z | [
"transformers",
"tf",
"bert",
"feature-extraction",
"generated_from_keras_callback",
"endpoints_compatible",
"region:us"
] | feature-extraction | ArthurZ | null | null | ArthurZ/tiny-random-bert-sharded | 0 | 7,569 | transformers | 2022-06-17T07:49:01 | ---
tags:
- generated_from_keras_callback
model-index:
- name: tiny-random-bert-sharded
results: []
---
<!-- This model card has been generated automatically according to the information Keras had access to. You should
probably proofread and complete it, then remove this comment. -->
# tiny-random-bert-sharded
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
## 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:
- optimizer: None
- training_precision: float32
### Training results
### Framework versions
- Transformers 4.21.0.dev0
- TensorFlow 2.9.0
- Datasets 2.2.2
- Tokenizers 0.12.1
| 887 | [
[
-0.0380859375,
-0.0458984375,
0.033416748046875,
-0.004665374755859375,
-0.047149658203125,
-0.03857421875,
-0.0034351348876953125,
-0.027313232421875,
0.01097869873046875,
0.03179931640625,
-0.04815673828125,
-0.028289794921875,
-0.0582275390625,
-0.029327392578125,
-0.029632568359375,
0.08740234375,
0.0267333984375,
0.04241943359375,
-0.0129547119140625,
-0.005626678466796875,
-0.021697998046875,
-0.03729248046875,
-0.0791015625,
-0.048614501953125,
0.0328369140625,
0.026611328125,
0.061614990234375,
0.0704345703125,
0.0556640625,
0.0261993408203125,
-0.013763427734375,
-0.0190887451171875,
-0.04022216796875,
-0.035400390625,
-0.0008831024169921875,
-0.036834716796875,
-0.033966064453125,
-0.0033283233642578125,
0.05712890625,
0.04022216796875,
-0.0205078125,
0.03753662109375,
-0.00423431396484375,
0.0235748291015625,
-0.032135009765625,
0.011016845703125,
-0.047210693359375,
0.042572021484375,
-0.01280975341796875,
-0.0002231597900390625,
-0.0207061767578125,
-0.0234832763671875,
0.0107879638671875,
-0.037078857421875,
0.04656982421875,
-0.0020160675048828125,
0.0926513671875,
0.0159149169921875,
-0.001949310302734375,
-0.0134124755859375,
-0.060577392578125,
0.0521240234375,
-0.049713134765625,
0.01235198974609375,
0.02691650390625,
0.0455322265625,
0.006885528564453125,
-0.0780029296875,
-0.035675048828125,
-0.0031528472900390625,
0.00771331787109375,
-0.004093170166015625,
-0.0232696533203125,
-0.006359100341796875,
0.036163330078125,
0.028289794921875,
-0.006076812744140625,
0.03271484375,
-0.051239013671875,
-0.0235595703125,
0.051300048828125,
0.0362548828125,
-0.0233306884765625,
-0.0252685546875,
-0.03521728515625,
-0.03741455078125,
-0.0169219970703125,
0.0051727294921875,
0.050201416015625,
0.021728515625,
-0.02789306640625,
0.054962158203125,
-0.00016117095947265625,
0.035369873046875,
0.0096893310546875,
0.011474609375,
0.0270843505859375,
0.01102447509765625,
-0.041015625,
-0.00048351287841796875,
0.067138671875,
0.0272369384765625,
0.003986358642578125,
-0.003940582275390625,
-0.034942626953125,
-0.0291748046875,
0.030181884765625,
-0.055450439453125,
-0.034332275390625,
0.01104736328125,
-0.07470703125,
-0.08282470703125,
-0.00043702125549316406,
-0.051422119140625,
-0.0144500732421875,
-0.018646240234375,
0.059600830078125,
-0.0211181640625,
-0.0010528564453125,
-0.00701904296875,
-0.0260162353515625,
0.0187225341796875,
0.01151275634765625,
-0.052520751953125,
0.0196990966796875,
0.04144287109375,
0.0251312255859375,
0.0101318359375,
-0.015899658203125,
0.0010967254638671875,
-0.007724761962890625,
-0.02734375,
0.033843994140625,
-0.0150146484375,
-0.035552978515625,
-0.021209716796875,
0.01154327392578125,
0.00830841064453125,
-0.028961181640625,
0.0743408203125,
-0.035430908203125,
0.0185089111328125,
-0.0399169921875,
-0.049468994140625,
-0.0219879150390625,
0.006866455078125,
-0.058013916015625,
0.080078125,
-0.00719451904296875,
-0.0498046875,
0.04718017578125,
-0.06109619140625,
-0.0157623291015625,
0.0147705078125,
-0.0079498291015625,
-0.06024169921875,
0.027587890625,
-0.004222869873046875,
0.042449951171875,
-0.0150604248046875,
-0.00089263916015625,
-0.033111572265625,
-0.036102294921875,
-0.00461578369140625,
-0.03741455078125,
0.058624267578125,
0.030853271484375,
-0.0242156982421875,
0.0008831024169921875,
-0.08062744140625,
0.0308990478515625,
0.036346435546875,
-0.0226287841796875,
-0.0008683204650878906,
-0.0168304443359375,
0.0333251953125,
0.0116424560546875,
0.03143310546875,
-0.057525634765625,
0.0027313232421875,
-0.0166778564453125,
0.0175933837890625,
0.06695556640625,
0.01380157470703125,
0.00389862060546875,
-0.0347900390625,
0.004528045654296875,
0.0201873779296875,
0.0150604248046875,
0.0247344970703125,
-0.04412841796875,
-0.07196044921875,
-0.0142974853515625,
0.04638671875,
0.006229400634765625,
-0.0160064697265625,
0.055572509765625,
-0.0009126663208007812,
-0.06805419921875,
-0.043792724609375,
0.01081085205078125,
0.02911376953125,
0.02789306640625,
0.0035724639892578125,
-0.017852783203125,
-0.04095458984375,
-0.08740234375,
0.02545166015625,
0.0002040863037109375,
-0.0004172325134277344,
0.0294342041015625,
0.05035400390625,
-0.015869140625,
0.04888916015625,
-0.042266845703125,
-0.01041412353515625,
-0.0068359375,
0.0168609619140625,
0.0234527587890625,
0.06829833984375,
0.037109375,
-0.0394287109375,
0.00604248046875,
-0.0167999267578125,
-0.045745849609375,
0.03302001953125,
-0.00977325439453125,
-0.0263519287109375,
-0.0194244384765625,
0.0245208740234375,
-0.04193115234375,
0.0305023193359375,
0.0126495361328125,
-0.0016069412231445312,
0.036102294921875,
-0.046905517578125,
-0.021026611328125,
-0.0762939453125,
0.0194549560546875,
-0.004306793212890625,
0.003826141357421875,
-0.0178985595703125,
0.0162506103515625,
-0.0009503364562988281,
-0.0299835205078125,
-0.031280517578125,
0.039276123046875,
0.00555419921875,
-0.0113525390625,
-0.01220703125,
-0.0282745361328125,
-0.004886627197265625,
0.044647216796875,
0.0206756591796875,
0.021453857421875,
0.0304718017578125,
-0.05426025390625,
0.032867431640625,
0.0299224853515625,
-0.036041259765625,
0.032012939453125,
-0.07183837890625,
0.025726318359375,
-0.0148162841796875,
-0.025299072265625,
-0.058258056640625,
-0.0294189453125,
0.0294647216796875,
-0.0233001708984375,
0.01204681396484375,
-0.0027790069580078125,
-0.049346923828125,
-0.046783447265625,
-0.004886627197265625,
0.022979736328125,
0.051513671875,
-0.061279296875,
0.029327392578125,
0.0156402587890625,
0.0321044921875,
-0.021148681640625,
-0.061920166015625,
-0.0269775390625,
0.0030517578125,
-0.0302276611328125,
0.0067596435546875,
-0.009765625,
-0.004146575927734375,
0.0093841552734375,
0.0208587646484375,
-0.029937744140625,
-0.005458831787109375,
0.0279541015625,
0.01617431640625,
-0.01617431640625,
0.01557159423828125,
0.00774383544921875,
-0.0033130645751953125,
0.0194549560546875,
0.01044464111328125,
0.034759521484375,
-0.0180816650390625,
-0.042144775390625,
-0.038116455078125,
-0.0087738037109375,
0.0271759033203125,
0.01202392578125,
0.04351806640625,
0.07012939453125,
-0.043243408203125,
-0.00795745849609375,
-0.033538818359375,
-0.015716552734375,
-0.0333251953125,
0.038848876953125,
-0.0307769775390625,
-0.0188751220703125,
0.0540771484375,
0.028900146484375,
0.01126861572265625,
0.06890869140625,
0.0543212890625,
-0.03179931640625,
0.07525634765625,
0.036041259765625,
0.00319671630859375,
0.01367950439453125,
-0.0404052734375,
-0.00972747802734375,
-0.043365478515625,
-0.037017822265625,
-0.031890869140625,
-0.040740966796875,
-0.053863525390625,
-0.0017528533935546875,
0.020263671875,
0.004962921142578125,
-0.0231781005859375,
0.029205322265625,
-0.0491943359375,
0.03179931640625,
0.06781005859375,
0.044647216796875,
-0.01389312744140625,
0.00759124755859375,
-0.033111572265625,
-0.007175445556640625,
-0.0679931640625,
-0.0178680419921875,
0.1043701171875,
0.051849365234375,
0.042694091796875,
-0.02630615234375,
0.052276611328125,
0.035491943359375,
0.0014467239379882812,
-0.037628173828125,
0.043853759765625,
0.00951385498046875,
-0.0716552734375,
-0.0262603759765625,
-0.0218353271484375,
-0.06793212890625,
0.00997161865234375,
-0.033660888671875,
-0.031524658203125,
0.0174407958984375,
0.004245758056640625,
-0.037628173828125,
0.03814697265625,
-0.046600341796875,
0.06939697265625,
-0.01180267333984375,
-0.006870269775390625,
-0.02508544921875,
-0.034088134765625,
0.032379150390625,
-0.0109100341796875,
-0.0191192626953125,
0.0005412101745605469,
0.0145416259765625,
0.08062744140625,
-0.059234619140625,
0.05364990234375,
-0.034881591796875,
0.024505615234375,
0.0391845703125,
-0.0188446044921875,
0.039703369140625,
-0.0036830902099609375,
-0.006256103515625,
0.0208740234375,
-0.002288818359375,
-0.0458984375,
-0.00620269775390625,
0.052093505859375,
-0.08758544921875,
-0.0001418590545654297,
-0.0247802734375,
-0.045074462890625,
0.0013275146484375,
0.019287109375,
0.047210693359375,
0.03778076171875,
-0.0216064453125,
0.021575927734375,
0.050079345703125,
0.004146575927734375,
0.0550537109375,
0.0236053466796875,
0.00933837890625,
-0.03643798828125,
0.054168701171875,
-0.0037250518798828125,
0.0006299018859863281,
-0.00920867919921875,
0.0170135498046875,
-0.018585205078125,
-0.032745361328125,
-0.0289306640625,
0.0081634521484375,
-0.0654296875,
-0.00928497314453125,
-0.0165252685546875,
-0.045562744140625,
-0.013946533203125,
-0.000988006591796875,
-0.0302886962890625,
-0.033477783203125,
-0.0205230712890625,
-0.0171966552734375,
0.010345458984375,
0.062042236328125,
-0.0230865478515625,
0.054168701171875,
-0.05731201171875,
0.007511138916015625,
0.041748046875,
0.033538818359375,
0.0267333984375,
-0.044036865234375,
-0.032257080078125,
0.00823211669921875,
-0.0229339599609375,
-0.0141448974609375,
0.006717681884765625,
-0.0048980712890625,
0.059356689453125,
0.05548095703125,
-0.0224761962890625,
0.044708251953125,
-0.0202789306640625,
0.054473876953125,
0.01148223876953125,
-0.034820556640625,
0.0180816650390625,
-0.01529693603515625,
0.0316162109375,
0.04522705078125,
0.0293731689453125,
0.01102447509765625,
-0.00685882568359375,
-0.107421875,
-0.028045654296875,
0.0533447265625,
0.033538818359375,
0.0244598388671875,
0.003803253173828125,
0.022979736328125,
0.006580352783203125,
0.0299072265625,
-0.055694580078125,
-0.035736083984375,
0.0035572052001953125,
-0.0012731552124023438,
0.006275177001953125,
-0.042572021484375,
-0.01468658447265625,
-0.046783447265625,
0.0743408203125,
0.028045654296875,
0.0531005859375,
-0.00809478759765625,
0.01019287109375,
-0.0271759033203125,
-0.0163726806640625,
0.0458984375,
0.038055419921875,
-0.04010009765625,
-0.0178680419921875,
0.01409912109375,
-0.045257568359375,
-0.0178375244140625,
0.01294708251953125,
-0.0196380615234375,
0.0214385986328125,
0.0150146484375,
0.06060791015625,
0.0279693603515625,
-0.0165252685546875,
0.03753662109375,
-0.004974365234375,
-0.02447509765625,
-0.0255279541015625,
0.0141448974609375,
-0.00954437255859375,
0.010406494140625,
0.0110321044921875,
0.054443359375,
-0.0027942657470703125,
-0.004192352294921875,
0.0299224853515625,
0.031280517578125,
-0.035064697265625,
-0.0134735107421875,
0.06304931640625,
0.00923919677734375,
-0.0235443115234375,
0.0552978515625,
-0.0159149169921875,
-0.01039886474609375,
0.0635986328125,
0.0216064453125,
0.06890869140625,
0.00011092424392700195,
0.00800323486328125,
0.0455322265625,
0.0187225341796875,
0.0011034011840820312,
0.03900146484375,
-0.01837158203125,
-0.06829833984375,
-0.004528045654296875,
-0.0546875,
-0.032318115234375,
0.0430908203125,
-0.081298828125,
0.048797607421875,
-0.06317138671875,
-0.0234527587890625,
0.033538818359375,
0.01143646240234375,
-0.0838623046875,
0.046112060546875,
0.0233001708984375,
0.0775146484375,
-0.07098388671875,
0.068359375,
0.03814697265625,
-0.033477783203125,
-0.052276611328125,
-0.0257415771484375,
-0.0243682861328125,
-0.09210205078125,
0.045257568359375,
-0.0135040283203125,
0.033905029296875,
0.0175933837890625,
-0.050018310546875,
-0.054779052734375,
0.07012939453125,
-0.001125335693359375,
-0.028656005859375,
-0.004238128662109375,
0.0250244140625,
0.041534423828125,
-0.013427734375,
0.0367431640625,
0.0184326171875,
0.0255279541015625,
0.01355743408203125,
-0.050567626953125,
-0.0196990966796875,
-0.015472412109375,
0.0182342529296875,
0.002338409423828125,
-0.0435791015625,
0.058929443359375,
-0.005603790283203125,
0.035369873046875,
0.0291900634765625,
0.042449951171875,
0.020263671875,
0.0011568069458007812,
0.039306640625,
0.079345703125,
0.050628662109375,
-0.001983642578125,
0.07708740234375,
-0.0275421142578125,
0.041534423828125,
0.0869140625,
0.01128387451171875,
0.047393798828125,
0.02496337890625,
-0.0140228271484375,
0.0196380615234375,
0.05950927734375,
-0.05987548828125,
0.051666259765625,
-0.00196075439453125,
0.00891876220703125,
-0.0235748291015625,
0.0157470703125,
-0.034454345703125,
0.037445068359375,
0.0005817413330078125,
-0.07586669921875,
-0.037841796875,
-0.006580352783203125,
-0.012908935546875,
-0.0261688232421875,
-0.049652099609375,
0.038055419921875,
-0.0180511474609375,
-0.011627197265625,
0.0241851806640625,
0.029205322265625,
0.0215911865234375,
-0.045684814453125,
-0.002681732177734375,
0.0167694091796875,
0.019683837890625,
-0.00330352783203125,
-0.049224853515625,
0.0088043212890625,
-0.01346588134765625,
-0.0311126708984375,
-0.0011110305786132812,
0.06036376953125,
-0.0068359375,
-0.07568359375,
-0.018890380859375,
0.0230560302734375,
0.033477783203125,
0.01444244384765625,
-0.07659912109375,
-0.005268096923828125,
-0.005420684814453125,
-0.02093505859375,
0.012847900390625,
0.0296173095703125,
0.01091766357421875,
0.0380859375,
0.04486083984375,
-0.0036830902099609375,
0.012298583984375,
0.00406646728515625,
0.055908203125,
-0.033050537109375,
-0.0528564453125,
-0.048980712890625,
0.03662109375,
-0.012603759765625,
-0.05426025390625,
0.052398681640625,
0.07781982421875,
0.06683349609375,
-0.027984619140625,
0.0430908203125,
-0.0160980224609375,
0.0296173095703125,
-0.025909423828125,
0.06390380859375,
-0.01468658447265625,
-0.00812530517578125,
-0.01033782958984375,
-0.0606689453125,
0.0213470458984375,
0.03851318359375,
-0.0130615234375,
0.01308441162109375,
0.028778076171875,
0.040283203125,
-0.0211181640625,
0.0198516845703125,
0.02972412109375,
0.0009522438049316406,
-0.0121917724609375,
0.013458251953125,
0.024658203125,
-0.06787109375,
0.039520263671875,
-0.06817626953125,
0.0180816650390625,
-0.00879669189453125,
-0.05853271484375,
-0.07183837890625,
-0.031494140625,
-0.0289306640625,
-0.026214599609375,
-0.01226043701171875,
0.06591796875,
0.0654296875,
-0.06549072265625,
-0.015716552734375,
-0.004299163818359375,
-0.01178741455078125,
-0.01192474365234375,
-0.0146026611328125,
0.038238525390625,
-0.0154571533203125,
-0.047393798828125,
-0.005741119384765625,
-0.034942626953125,
0.024566650390625,
-0.0199432373046875,
-0.0072174072265625,
0.0031261444091796875,
-0.00972747802734375,
0.00266265869140625,
0.0011758804321289062,
-0.031036376953125,
-0.041168212890625,
-0.033294677734375,
-0.01904296875,
0.013214111328125,
0.0235443115234375,
-0.04095458984375,
0.036865234375,
0.0185089111328125,
0.031219482421875,
0.06256103515625,
-0.004913330078125,
0.03192138671875,
-0.062469482421875,
0.033355712890625,
0.0143890380859375,
0.02813720703125,
-0.006221771240234375,
-0.040008544921875,
0.0278472900390625,
0.0216827392578125,
-0.0296173095703125,
-0.06402587890625,
-0.0223236083984375,
-0.062164306640625,
0.016815185546875,
0.06744384765625,
0.0096893310546875,
-0.035614013671875,
0.04205322265625,
-0.00019419193267822266,
0.0108489990234375,
-0.01338958740234375,
0.040313720703125,
0.057037353515625,
0.0019235610961914062,
0.00643157958984375,
-0.02911376953125,
0.0296173095703125,
0.0266265869140625,
-0.031341552734375,
-0.041595458984375,
0.009857177734375,
0.040496826171875,
0.005615234375,
0.01361083984375,
-0.0160675048828125,
0.03753662109375,
0.0251007080078125,
0.035675048828125,
-0.03729248046875,
-0.01264190673828125,
-0.032257080078125,
0.018646240234375,
0.0031261444091796875,
-0.055328369140625
]
] |
facebook/dinov2-base-imagenet1k-1-layer | 2023-09-15T06:40:46.000Z | [
"transformers",
"pytorch",
"dinov2",
"image-classification",
"dino",
"vision",
"dataset:imagenet-1k",
"arxiv:2304.07193",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | image-classification | facebook | null | null | facebook/dinov2-base-imagenet1k-1-layer | 0 | 7,563 | transformers | 2023-09-14T19:59:55 | ---
license: apache-2.0
tags:
- dino
- vision
datasets:
- imagenet-1k
---
# Vision Transformer (base-sized model) trained using DINOv2
Vision Transformer (ViT) model trained using the DINOv2 method. It was introduced in the paper [DINOv2: Learning Robust Visual Features without Supervision](https://arxiv.org/abs/2304.07193) by Oquab et al. and first released in [this repository](https://github.com/facebookresearch/dinov2).
Disclaimer: The team releasing DINOv2 did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a self-supervised fashion.
Images are presented to the model as a sequence of fixed-size patches, which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
Note that this model does not include any fine-tuned heads.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the model for classifying an image among one of the [1000 ImageNet labels](https://huggingface.co/datasets/huggingface/label-files/blob/main/imagenet-1k-id2label.json). See the [model hub](https://huggingface.co/models?search=facebook/dinov2) to look for
other fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import AutoImageProcessor, AutoModelForImageClassification
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = AutoImageProcessor.from_pretrained('facebook/dinov2-base-imagenet1k-1-layer')
model = AutoModelForImageClassification.from_pretrained('facebook/dinov2-base-imagenet1k-1-layer')
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
### BibTeX entry and citation info
```bibtex
misc{oquab2023dinov2,
title={DINOv2: Learning Robust Visual Features without Supervision},
author={Maxime Oquab and Timothée Darcet and Théo Moutakanni and Huy Vo and Marc Szafraniec and Vasil Khalidov and Pierre Fernandez and Daniel Haziza and Francisco Massa and Alaaeldin El-Nouby and Mahmoud Assran and Nicolas Ballas and Wojciech Galuba and Russell Howes and Po-Yao Huang and Shang-Wen Li and Ishan Misra and Michael Rabbat and Vasu Sharma and Gabriel Synnaeve and Hu Xu and Hervé Jegou and Julien Mairal and Patrick Labatut and Armand Joulin and Piotr Bojanowski},
year={2023},
eprint={2304.07193},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
``` | 3,364 | [
[
-0.042144775390625,
-0.028564453125,
-0.002849578857421875,
-0.00823974609375,
-0.0304107666015625,
-0.00811004638671875,
0.0107269287109375,
-0.034271240234375,
0.0216522216796875,
0.035614013671875,
-0.032379150390625,
-0.0166168212890625,
-0.05670166015625,
-0.01523590087890625,
-0.028411865234375,
0.0704345703125,
-0.0006289482116699219,
-0.0005578994750976562,
-0.027435302734375,
-0.00818634033203125,
-0.0172271728515625,
-0.035247802734375,
-0.051513671875,
-0.0357666015625,
0.037017822265625,
0.007843017578125,
0.053497314453125,
0.07586669921875,
0.033538818359375,
0.03094482421875,
-0.01242828369140625,
-0.0007443428039550781,
-0.03973388671875,
-0.0191802978515625,
-0.01171875,
-0.03857421875,
-0.0261688232421875,
0.01068878173828125,
0.039093017578125,
0.033599853515625,
0.018310546875,
0.022216796875,
0.00946044921875,
0.0171051025390625,
-0.0411376953125,
0.03131103515625,
-0.0242767333984375,
0.0276031494140625,
-0.007328033447265625,
-0.0010395050048828125,
-0.025787353515625,
-0.0228729248046875,
0.0208740234375,
-0.03643798828125,
0.0049591064453125,
0.00016570091247558594,
0.0931396484375,
0.021881103515625,
-0.035614013671875,
-0.0027256011962890625,
-0.03814697265625,
0.0570068359375,
-0.020751953125,
0.025146484375,
0.02044677734375,
0.0273284912109375,
0.00798797607421875,
-0.07977294921875,
-0.04150390625,
0.0039825439453125,
-0.01216888427734375,
0.006549835205078125,
-0.01666259765625,
0.00354766845703125,
0.022705078125,
0.025360107421875,
-0.016815185546875,
0.015960693359375,
-0.039886474609375,
-0.028045654296875,
0.0278472900390625,
-0.01043701171875,
0.014923095703125,
-0.02520751953125,
-0.058074951171875,
-0.0281829833984375,
-0.024139404296875,
0.0298309326171875,
0.0161895751953125,
0.005702972412109375,
-0.01458740234375,
0.0433349609375,
0.00417327880859375,
0.044921875,
0.02587890625,
-0.012542724609375,
0.041656494140625,
-0.0186004638671875,
-0.0232391357421875,
-0.002925872802734375,
0.062744140625,
0.0254669189453125,
0.0248870849609375,
0.003398895263671875,
-0.0234527587890625,
0.006378173828125,
0.0208892822265625,
-0.07421875,
-0.0229034423828125,
-0.006259918212890625,
-0.0443115234375,
-0.042388916015625,
0.0156402587890625,
-0.048614501953125,
-0.01192474365234375,
-0.026580810546875,
0.053497314453125,
-0.0201263427734375,
-0.0255889892578125,
-0.0276031494140625,
-0.0003857612609863281,
0.051544189453125,
0.009246826171875,
-0.05987548828125,
0.029083251953125,
0.0340576171875,
0.0687255859375,
-0.005580902099609375,
-0.018707275390625,
-0.01739501953125,
-0.0129852294921875,
-0.036895751953125,
0.047393798828125,
-0.0208282470703125,
-0.0096588134765625,
0.0093994140625,
0.03369140625,
-0.001617431640625,
-0.032196044921875,
0.0236358642578125,
-0.0303497314453125,
0.018798828125,
-0.021392822265625,
-0.0286712646484375,
-0.0231170654296875,
0.01561737060546875,
-0.0447998046875,
0.08123779296875,
0.0253753662109375,
-0.0673828125,
0.036376953125,
-0.03338623046875,
-0.014801025390625,
0.0029850006103515625,
-0.0038928985595703125,
-0.055206298828125,
-0.0172576904296875,
0.0172119140625,
0.03863525390625,
0.0055389404296875,
-0.00786590576171875,
-0.0299072265625,
-0.034271240234375,
0.0192413330078125,
-0.00458526611328125,
0.07415771484375,
0.00902557373046875,
-0.0291900634765625,
0.01007843017578125,
-0.04827880859375,
0.00243377685546875,
0.01800537109375,
-0.0223846435546875,
-0.0119476318359375,
-0.02099609375,
0.01334381103515625,
0.0264129638671875,
0.0202178955078125,
-0.04595947265625,
0.018310546875,
-0.0188751220703125,
0.04217529296875,
0.05987548828125,
-0.00525665283203125,
0.04229736328125,
-0.018157958984375,
0.028839111328125,
0.01123046875,
0.039581298828125,
-0.0256805419921875,
-0.048126220703125,
-0.06707763671875,
-0.0146331787109375,
0.02471923828125,
0.03948974609375,
-0.06756591796875,
0.03802490234375,
-0.011962890625,
-0.0256195068359375,
-0.0298309326171875,
0.00951385498046875,
0.04327392578125,
0.04693603515625,
0.0282135009765625,
-0.04229736328125,
-0.044158935546875,
-0.06854248046875,
0.01617431640625,
-0.006328582763671875,
0.006961822509765625,
0.02154541015625,
0.051025390625,
-0.0254669189453125,
0.06622314453125,
-0.00771331787109375,
-0.01898193359375,
0.0022144317626953125,
0.0037517547607421875,
0.0130615234375,
0.05718994140625,
0.059173583984375,
-0.0701904296875,
-0.02593994140625,
-0.008636474609375,
-0.0655517578125,
0.01175689697265625,
0.00540924072265625,
-0.0218048095703125,
0.003833770751953125,
0.0262908935546875,
-0.051910400390625,
0.05657958984375,
0.01361083984375,
-0.0198974609375,
0.0166015625,
-0.0031490325927734375,
0.0006051063537597656,
-0.0850830078125,
-0.0028362274169921875,
0.000052034854888916016,
-0.0318603515625,
-0.046051025390625,
0.0091705322265625,
0.00733184814453125,
-0.004711151123046875,
-0.03759765625,
0.032379150390625,
-0.035552978515625,
-0.0229034423828125,
-0.0215606689453125,
-0.0192413330078125,
-0.0004954338073730469,
0.04266357421875,
-0.00037598609924316406,
0.0271148681640625,
0.0687255859375,
-0.03350830078125,
0.053253173828125,
0.032745361328125,
-0.032257080078125,
0.04327392578125,
-0.055999755859375,
0.0246429443359375,
-0.0193939208984375,
0.0166168212890625,
-0.07568359375,
-0.032928466796875,
0.036102294921875,
-0.0372314453125,
0.04937744140625,
-0.0281524658203125,
-0.03204345703125,
-0.059967041015625,
-0.0170745849609375,
0.0325927734375,
0.056915283203125,
-0.05889892578125,
0.040069580078125,
0.0261383056640625,
0.0233154296875,
-0.06378173828125,
-0.08489990234375,
-0.00771331787109375,
-0.01232147216796875,
-0.035675048828125,
0.0255889892578125,
0.016998291015625,
0.0184478759765625,
0.0265350341796875,
-0.00678253173828125,
-0.01953125,
-0.0182037353515625,
0.048614501953125,
0.0236053466796875,
-0.026031494140625,
0.0018968582153320312,
-0.014739990234375,
-0.010955810546875,
0.00669097900390625,
-0.046295166015625,
0.036102294921875,
-0.019500732421875,
-0.0233917236328125,
-0.054595947265625,
0.00502777099609375,
0.048614501953125,
-0.0186004638671875,
0.051513671875,
0.072509765625,
-0.050018310546875,
-0.00945281982421875,
-0.02496337890625,
-0.0113067626953125,
-0.03924560546875,
0.023651123046875,
-0.02423095703125,
-0.04443359375,
0.05682373046875,
0.00226593017578125,
-0.019317626953125,
0.03460693359375,
0.03546142578125,
-0.0105743408203125,
0.06341552734375,
0.05706787109375,
0.0020236968994140625,
0.054229736328125,
-0.057159423828125,
0.0009322166442871094,
-0.05841064453125,
-0.052734375,
-0.002704620361328125,
-0.033111572265625,
-0.032928466796875,
-0.03582763671875,
0.00859832763671875,
0.0293121337890625,
-0.0257568359375,
0.04443359375,
-0.055023193359375,
0.032867431640625,
0.0615234375,
0.040283203125,
-0.0267486572265625,
0.0087127685546875,
-0.0206298828125,
0.00972747802734375,
-0.0504150390625,
-0.0070953369140625,
0.07354736328125,
0.044219970703125,
0.057708740234375,
-0.00965118408203125,
0.044891357421875,
0.007434844970703125,
0.0090789794921875,
-0.069091796875,
0.0360107421875,
-0.01049041748046875,
-0.042724609375,
-0.010223388671875,
-0.0193328857421875,
-0.0704345703125,
-0.00678253173828125,
-0.0245208740234375,
-0.056304931640625,
0.04730224609375,
0.0206451416015625,
-0.024444580078125,
0.02752685546875,
-0.04541015625,
0.0723876953125,
-0.0177001953125,
-0.02471923828125,
0.01279449462890625,
-0.05364990234375,
0.0161285400390625,
-0.004749298095703125,
-0.01145172119140625,
0.0208740234375,
0.0149688720703125,
0.0565185546875,
-0.046630859375,
0.08087158203125,
-0.03363037109375,
0.0247039794921875,
0.047393798828125,
-0.01239776611328125,
0.027191162109375,
-0.00821685791015625,
0.027008056640625,
0.0203704833984375,
-0.0024566650390625,
-0.040252685546875,
-0.038330078125,
0.036773681640625,
-0.07830810546875,
-0.0248565673828125,
-0.0308685302734375,
-0.01488494873046875,
0.0189208984375,
0.02862548828125,
0.053985595703125,
0.0546875,
0.014892578125,
0.033843994140625,
0.050933837890625,
-0.025390625,
0.033935546875,
-0.01824951171875,
-0.0274200439453125,
-0.0286865234375,
0.061126708984375,
0.0269622802734375,
0.01873779296875,
0.022491455078125,
0.0145111083984375,
-0.0280303955078125,
-0.0262451171875,
-0.0265960693359375,
0.0014019012451171875,
-0.07452392578125,
-0.0261383056640625,
-0.036285400390625,
-0.046234130859375,
-0.036163330078125,
-0.01064300537109375,
-0.040283203125,
-0.027679443359375,
-0.035430908203125,
-0.016387939453125,
0.023773193359375,
0.06304931640625,
-0.02252197265625,
0.039642333984375,
-0.0243072509765625,
0.01493072509765625,
0.058349609375,
0.0287017822265625,
-0.0010385513305664062,
-0.050689697265625,
-0.01580810546875,
0.0034198760986328125,
-0.006160736083984375,
-0.0504150390625,
0.027587890625,
0.0273590087890625,
0.056488037109375,
0.05657958984375,
-0.023406982421875,
0.05560302734375,
-0.0228729248046875,
0.04730224609375,
0.033111572265625,
-0.058502197265625,
0.046783447265625,
-0.015045166015625,
0.01702880859375,
0.0130157470703125,
0.043212890625,
-0.004604339599609375,
0.0175018310546875,
-0.036407470703125,
-0.04888916015625,
0.061279296875,
0.0158843994140625,
0.0171051025390625,
0.004180908203125,
0.05059814453125,
-0.0072784423828125,
0.0038776397705078125,
-0.0679931640625,
-0.0243072509765625,
-0.0716552734375,
-0.005931854248046875,
0.01064300537109375,
-0.0255584716796875,
-0.0060577392578125,
-0.040496826171875,
0.0250396728515625,
-0.0057830810546875,
0.055267333984375,
0.01509857177734375,
-0.00959014892578125,
-0.015716552734375,
-0.0364990234375,
0.01396942138671875,
0.0325927734375,
-0.0271148681640625,
0.0171966552734375,
0.005451202392578125,
-0.03857421875,
-0.000025570392608642578,
0.005474090576171875,
-0.01404571533203125,
-0.007114410400390625,
0.03485107421875,
0.0762939453125,
0.01255035400390625,
0.00048160552978515625,
0.0694580078125,
0.00791168212890625,
-0.022216796875,
-0.039825439453125,
0.00891876220703125,
-0.00771331787109375,
0.03466796875,
0.02130126953125,
0.032684326171875,
-0.0015344619750976562,
-0.0477294921875,
0.038360595703125,
0.0191192626953125,
-0.05157470703125,
-0.034271240234375,
0.06353759765625,
-0.00910186767578125,
-0.0146636962890625,
0.046722412109375,
-0.01554107666015625,
-0.052734375,
0.0673828125,
0.04547119140625,
0.05059814453125,
-0.027984619140625,
0.0194091796875,
0.044158935546875,
0.0183258056640625,
-0.0084228515625,
0.00974273681640625,
-0.0095672607421875,
-0.06793212890625,
-0.02801513671875,
-0.050537109375,
-0.0030956268310546875,
0.01007843017578125,
-0.06390380859375,
0.0225372314453125,
-0.057586669921875,
-0.03253173828125,
0.022247314453125,
-0.0158538818359375,
-0.0802001953125,
0.022796630859375,
0.04022216796875,
0.05255126953125,
-0.056884765625,
0.07763671875,
0.060546875,
-0.0390625,
-0.05657958984375,
-0.025604248046875,
0.00511932373046875,
-0.0751953125,
0.06793212890625,
0.0306396484375,
0.0006122589111328125,
0.002655029296875,
-0.06976318359375,
-0.07916259765625,
0.08831787109375,
0.0161590576171875,
-0.01605224609375,
-0.003910064697265625,
0.004146575927734375,
0.03338623046875,
-0.040313720703125,
0.0260772705078125,
-0.002353668212890625,
0.01329803466796875,
0.039459228515625,
-0.05633544921875,
-0.0010013580322265625,
-0.03076171875,
0.0231475830078125,
-0.0070953369140625,
-0.062744140625,
0.084228515625,
-0.01244354248046875,
-0.01177215576171875,
0.01031494140625,
0.04254150390625,
-0.0226898193359375,
0.002216339111328125,
0.047637939453125,
0.049530029296875,
0.04205322265625,
-0.021270751953125,
0.0748291015625,
-0.00455474853515625,
0.045562744140625,
0.055206298828125,
0.0208740234375,
0.0450439453125,
0.0185394287109375,
-0.00568389892578125,
0.0472412109375,
0.06658935546875,
-0.03594970703125,
0.059844970703125,
0.0054779052734375,
0.0095672607421875,
-0.022186279296875,
-0.0038318634033203125,
-0.0294952392578125,
0.04901123046875,
0.0290374755859375,
-0.0462646484375,
0.0018787384033203125,
0.0216064453125,
-0.01378631591796875,
-0.0262908935546875,
-0.03497314453125,
0.047454833984375,
0.009246826171875,
-0.033477783203125,
0.0550537109375,
-0.018768310546875,
0.04254150390625,
-0.030975341796875,
-0.007053375244140625,
-0.009429931640625,
0.022369384765625,
-0.025360107421875,
-0.0650634765625,
0.0105743408203125,
-0.0177459716796875,
-0.00455474853515625,
-0.00832366943359375,
0.06719970703125,
-0.0251312255859375,
-0.042388916015625,
0.0305633544921875,
0.00817108154296875,
0.019073486328125,
0.018707275390625,
-0.0634765625,
-0.0148468017578125,
-0.00749969482421875,
-0.034942626953125,
0.01800537109375,
0.029205322265625,
0.006038665771484375,
0.05059814453125,
0.04473876953125,
-0.01520538330078125,
0.02886962890625,
0.00038242340087890625,
0.0848388671875,
-0.030517578125,
-0.031982421875,
-0.044952392578125,
0.042572021484375,
-0.0150909423828125,
-0.0271148681640625,
0.04266357421875,
0.023040771484375,
0.07635498046875,
-0.00556182861328125,
0.03662109375,
-0.01239013671875,
0.01247406005859375,
-0.02587890625,
0.04754638671875,
-0.034271240234375,
-0.016204833984375,
-0.0133056640625,
-0.07794189453125,
-0.0243072509765625,
0.07025146484375,
-0.0025787353515625,
0.00836181640625,
0.0312347412109375,
0.053009033203125,
-0.0225982666015625,
-0.0217132568359375,
0.0164947509765625,
0.0277099609375,
-0.00199127197265625,
0.03143310546875,
0.060302734375,
-0.04376220703125,
0.040313720703125,
-0.049530029296875,
-0.0290679931640625,
-0.011474609375,
-0.048553466796875,
-0.1002197265625,
-0.046051025390625,
-0.031280517578125,
-0.043304443359375,
0.0028591156005859375,
0.0567626953125,
0.0892333984375,
-0.0732421875,
0.01239776611328125,
-0.0057830810546875,
-0.00644683837890625,
-0.0157318115234375,
-0.013763427734375,
0.036407470703125,
-0.0032501220703125,
-0.05133056640625,
-0.0017518997192382812,
0.0061492919921875,
0.023223876953125,
-0.0261993408203125,
-0.004055023193359375,
-0.006259918212890625,
-0.01396942138671875,
0.040557861328125,
0.0269927978515625,
-0.052825927734375,
-0.04559326171875,
-0.0042572021484375,
-0.0046539306640625,
0.023529052734375,
0.029693603515625,
-0.06561279296875,
0.052154541015625,
0.03521728515625,
0.040435791015625,
0.064208984375,
0.005126953125,
0.01557159423828125,
-0.0587158203125,
0.0266265869140625,
0.00336456298828125,
0.039306640625,
0.0272674560546875,
-0.028594970703125,
0.03314208984375,
0.032379150390625,
-0.040557861328125,
-0.05487060546875,
0.01568603515625,
-0.08819580078125,
-0.01309967041015625,
0.0726318359375,
-0.038360595703125,
-0.039306640625,
0.00722503662109375,
-0.0013675689697265625,
0.04095458984375,
-0.0027866363525390625,
0.03741455078125,
0.0247955322265625,
0.004047393798828125,
-0.04644775390625,
-0.0290069580078125,
0.032867431640625,
-0.007904052734375,
-0.0290374755859375,
-0.0457763671875,
0.002124786376953125,
0.028076171875,
0.0290985107421875,
0.01378631591796875,
-0.0218505859375,
0.0123443603515625,
0.027008056640625,
0.0170135498046875,
-0.0206146240234375,
-0.0278472900390625,
-0.0162506103515625,
0.004863739013671875,
-0.0249481201171875,
-0.05072021484375
]
] |
Yntec/Photosphere | 2023-07-14T23:22:58.000Z | [
"diffusers",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"Noosphere",
"Dreamlike",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | Yntec | null | null | Yntec/Photosphere | 2 | 7,562 | diffusers | 2023-07-14T22:54:19 | ---
license: creativeml-openrail-m
library_name: diffusers
pipeline_tag: text-to-image
tags:
- stable-diffusion
- stable-diffusion-diffusers
- diffusers
- text-to-image
- Noosphere
- Dreamlike
---
# Photosphere
A mix of Noosphere v3 by skumerz and photorealistic models.
Original page:
https://civitai.com/models/36538?modelVersionId=107675 | 343 | [
[
-0.029266357421875,
-0.0137786865234375,
0.04376220703125,
0.041229248046875,
-0.024871826171875,
0.0158233642578125,
0.04071044921875,
-0.035552978515625,
0.045318603515625,
0.06610107421875,
-0.07940673828125,
-0.036834716796875,
0.0052032470703125,
-0.0283355712890625,
-0.02740478515625,
0.032135009765625,
0.00244140625,
0.020294189453125,
-0.0163116455078125,
-0.00521087646484375,
0.0032176971435546875,
-0.01461029052734375,
-0.010528564453125,
-0.035247802734375,
0.040496826171875,
0.064453125,
0.0438232421875,
-0.0084075927734375,
0.03228759765625,
0.01012420654296875,
-0.01195526123046875,
-0.021209716796875,
-0.01934814453125,
0.0000017285346984863281,
-0.01207733154296875,
-0.01666259765625,
-0.036834716796875,
0.007503509521484375,
0.0271453857421875,
0.018310546875,
-0.0206451416015625,
0.031097412109375,
-0.029937744140625,
0.019927978515625,
-0.045166015625,
-0.01690673828125,
-0.014617919921875,
0.0250244140625,
-0.0080718994140625,
-0.011505126953125,
0.0066680908203125,
-0.007083892822265625,
-0.01338958740234375,
-0.08154296875,
0.016510009765625,
0.00014138221740722656,
0.061370849609375,
-0.0018224716186523438,
-0.0352783203125,
0.0214691162109375,
-0.06829833984375,
0.048583984375,
-0.037261962890625,
0.0428466796875,
-0.0103759765625,
0.058746337890625,
-0.0313720703125,
-0.0513916015625,
-0.0312347412109375,
0.01184844970703125,
0.03662109375,
0.040191650390625,
-0.01107025146484375,
-0.01165008544921875,
0.0108795166015625,
0.0377197265625,
-0.04681396484375,
-0.0017375946044921875,
-0.0574951171875,
0.0089569091796875,
0.0311431884765625,
0.01248931884765625,
0.0289306640625,
-0.023651123046875,
-0.043121337890625,
0.0131072998046875,
-0.0309600830078125,
-0.0027942657470703125,
0.039703369140625,
-0.01390838623046875,
-0.01526641845703125,
0.0684814453125,
-0.03717041015625,
0.06317138671875,
-0.01360321044921875,
-0.0030727386474609375,
0.026275634765625,
-0.00928497314453125,
-0.0275421142578125,
-0.0203399658203125,
0.0197906494140625,
0.0660400390625,
-0.0238800048828125,
-0.019439697265625,
-0.006561279296875,
0.007450103759765625,
0.00928497314453125,
-0.07037353515625,
-0.033599853515625,
0.027557373046875,
-0.021575927734375,
-0.0545654296875,
0.0283050537109375,
-0.051483154296875,
-0.01557159423828125,
-0.0190582275390625,
0.00787353515625,
-0.01474761962890625,
-0.042877197265625,
0.027008056640625,
-0.0182342529296875,
0.016143798828125,
0.0248870849609375,
-0.031524658203125,
0.047149658203125,
0.01953125,
0.04962158203125,
0.044464111328125,
-0.005146026611328125,
0.01380157470703125,
0.004077911376953125,
-0.0271759033203125,
0.071533203125,
-0.00531005859375,
-0.049774169921875,
-0.0174102783203125,
0.01174163818359375,
0.01151275634765625,
-0.0140838623046875,
0.036376953125,
-0.056182861328125,
0.0101776123046875,
-0.02117919921875,
-0.018829345703125,
-0.0308074951171875,
0.007785797119140625,
-0.0496826171875,
0.09503173828125,
0.016571044921875,
-0.060577392578125,
0.0413818359375,
-0.05462646484375,
0.0114898681640625,
0.039642333984375,
-0.006053924560546875,
-0.021148681640625,
0.036712646484375,
-0.0142059326171875,
-0.01255035400390625,
-0.0031337738037109375,
-0.0213775634765625,
-0.04791259765625,
-0.025604248046875,
-0.01666259765625,
0.0245513916015625,
0.06915283203125,
0.036163330078125,
-0.0100555419921875,
0.03485107421875,
-0.071044921875,
0.0262298583984375,
0.0194244384765625,
0.0240936279296875,
-0.00972747802734375,
-0.01531982421875,
0.048370361328125,
0.05059814453125,
0.01776123046875,
-0.05389404296875,
0.019622802734375,
-0.008819580078125,
0.0177001953125,
0.027740478515625,
0.039764404296875,
0.036773681640625,
-0.032684326171875,
0.05657958984375,
0.01139068603515625,
0.05194091796875,
0.021728515625,
-0.0205230712890625,
-0.04083251953125,
-0.0183563232421875,
-0.00388336181640625,
0.033111572265625,
-0.07269287109375,
-0.0031185150146484375,
0.0024967193603515625,
-0.087646484375,
0.01013946533203125,
0.0221710205078125,
-0.012908935546875,
0.003055572509765625,
0.00502777099609375,
0.0004658699035644531,
-0.04168701171875,
-0.08831787109375,
-0.00370025634765625,
0.0166168212890625,
-0.0100250244140625,
0.046478271484375,
0.055816650390625,
-0.0083770751953125,
0.04144287109375,
-0.032257080078125,
0.0017528533935546875,
-0.01537322998046875,
-0.03131103515625,
0.041046142578125,
0.019256591796875,
0.0860595703125,
-0.045623779296875,
-0.045684814453125,
-0.0010814666748046875,
-0.038330078125,
-0.00620269775390625,
0.033477783203125,
-0.0305633544921875,
-0.00971221923828125,
0.038330078125,
-0.052215576171875,
0.018157958984375,
0.037567138671875,
-0.038116455078125,
0.0377197265625,
-0.0160369873046875,
0.0229644775390625,
-0.07257080078125,
0.0138702392578125,
0.01788330078125,
-0.02862548828125,
-0.027557373046875,
0.032470703125,
0.005405426025390625,
-0.0261383056640625,
-0.07598876953125,
0.036773681640625,
-0.055389404296875,
0.00801849365234375,
-0.0141754150390625,
-0.002407073974609375,
0.01213836669921875,
0.005245208740234375,
0.00484466552734375,
0.0301666259765625,
0.068603515625,
-0.0081329345703125,
0.032501220703125,
0.0075225830078125,
-0.037384033203125,
0.045166015625,
-0.06573486328125,
0.00629425048828125,
-0.0145111083984375,
0.01119232177734375,
-0.0574951171875,
-0.0555419921875,
0.0241851806640625,
-0.006061553955078125,
-0.047821044921875,
0.00408172607421875,
-0.07110595703125,
-0.04815673828125,
-0.016143798828125,
0.033355712890625,
0.034393310546875,
-0.0447998046875,
0.024810791015625,
0.010955810546875,
0.0187225341796875,
0.0004992485046386719,
-0.01806640625,
0.01399993896484375,
-0.01314544677734375,
-0.039947509765625,
0.0555419921875,
0.0021877288818359375,
-0.022979736328125,
0.023162841796875,
-0.01030731201171875,
-0.0104217529296875,
-0.02001953125,
0.041961669921875,
0.047760009765625,
-0.0272674560546875,
-0.03460693359375,
-0.01258087158203125,
0.0228424072265625,
-0.0264892578125,
-0.016845703125,
0.038330078125,
-0.0263519287109375,
0.006023406982421875,
-0.06060791015625,
0.0294189453125,
0.0831298828125,
0.0279083251953125,
0.05670166015625,
0.010223388671875,
-0.04803466796875,
0.0022449493408203125,
-0.039276123046875,
-0.006542205810546875,
-0.0278167724609375,
-0.0157928466796875,
-0.036865234375,
-0.04437255859375,
0.043365478515625,
0.018096923828125,
-0.0131072998046875,
0.0419921875,
0.0036869049072265625,
-0.01035308837890625,
0.0804443359375,
0.036102294921875,
0.01306915283203125,
0.04644775390625,
-0.049224853515625,
-0.0047454833984375,
-0.05023193359375,
-0.042022705078125,
-0.00954437255859375,
-0.0116119384765625,
-0.0236358642578125,
-0.0455322265625,
0.004978179931640625,
0.020965576171875,
-0.034515380859375,
0.03155517578125,
-0.0248870849609375,
0.06463623046875,
0.035186767578125,
0.0158233642578125,
0.01983642578125,
0.006618499755859375,
0.0026187896728515625,
-0.00862884521484375,
-0.059356689453125,
-0.025634765625,
0.03851318359375,
-0.013275146484375,
0.0299072265625,
0.00510406494140625,
0.047271728515625,
-0.0140228271484375,
0.02093505859375,
-0.022247314453125,
0.04034423828125,
0.0184173583984375,
-0.082275390625,
0.031097412109375,
-0.00948333740234375,
-0.0284881591796875,
0.04827880859375,
-0.050689697265625,
-0.032928466796875,
0.02484130859375,
-0.0008053779602050781,
-0.039703369140625,
0.027099609375,
-0.0272064208984375,
0.0645751953125,
-0.01018524169921875,
-0.012939453125,
0.00424957275390625,
-0.025146484375,
0.034759521484375,
0.043548583984375,
0.021026611328125,
-0.0082855224609375,
-0.0089569091796875,
0.01519775390625,
-0.0292205810546875,
0.045135498046875,
-0.02056884765625,
0.026702880859375,
0.028350830078125,
-0.0036163330078125,
-0.01490020751953125,
0.025299072265625,
-0.00966644287109375,
0.007534027099609375,
-0.0017137527465820312,
-0.06536865234375,
-0.027313232421875,
0.077392578125,
-0.07275390625,
-0.01050567626953125,
-0.052337646484375,
-0.006320953369140625,
0.0120697021484375,
0.05059814453125,
0.06280517578125,
0.0350341796875,
-0.057861328125,
0.0207366943359375,
0.0033016204833984375,
-0.0023975372314453125,
0.0172576904296875,
0.0264739990234375,
-0.073486328125,
-0.060546875,
0.05255126953125,
0.01302337646484375,
0.0223541259765625,
-0.0221099853515625,
-0.007720947265625,
-0.004978179931640625,
-0.03631591796875,
-0.01218414306640625,
0.0245819091796875,
-0.02716064453125,
-0.0211181640625,
-0.022674560546875,
-0.0367431640625,
-0.03460693359375,
-0.0275115966796875,
-0.02935791015625,
-0.05194091796875,
-0.045654296875,
-0.0190887451171875,
0.027496337890625,
0.070068359375,
0.01451873779296875,
0.056976318359375,
-0.020904541015625,
0.0328369140625,
0.004741668701171875,
0.0341796875,
-0.037567138671875,
-0.0099029541015625,
0.01284027099609375,
-0.0106964111328125,
-0.05126953125,
-0.034027099609375,
0.04388427734375,
-0.002429962158203125,
0.04730224609375,
0.0052337646484375,
-0.0230865478515625,
0.043121337890625,
-0.0183563232421875,
0.064697265625,
0.041229248046875,
-0.061676025390625,
0.049774169921875,
-0.0662841796875,
0.03564453125,
0.048858642578125,
0.0011243820190429688,
-0.0301055908203125,
-0.031646728515625,
-0.11151123046875,
-0.08526611328125,
0.03900146484375,
0.0291748046875,
-0.0126495361328125,
0.0201873779296875,
0.03277587890625,
-0.0005807876586914062,
0.044219970703125,
-0.0745849609375,
-0.01043701171875,
-0.0504150390625,
0.0005507469177246094,
-0.0165252685546875,
-0.01522064208984375,
-0.01345062255859375,
-0.025177001953125,
0.0467529296875,
0.00798797607421875,
-0.01360321044921875,
0.03936767578125,
0.00922393798828125,
-0.0224456787109375,
-0.01531982421875,
0.048614501953125,
0.0404052734375,
-0.04156494140625,
-0.0019931793212890625,
-0.0179901123046875,
0.004688262939453125,
0.0065765380859375,
-0.02862548828125,
-0.01160430908203125,
0.0022945404052734375,
0.00009357929229736328,
0.061614990234375,
0.038299560546875,
-0.01032257080078125,
0.055755615234375,
0.0077667236328125,
-0.03826904296875,
-0.07220458984375,
0.0260009765625,
0.007175445556640625,
0.02923583984375,
0.003940582275390625,
0.059722900390625,
0.050872802734375,
-0.0247039794921875,
0.0014133453369140625,
0.0163116455078125,
-0.03631591796875,
-0.052276611328125,
0.07427978515625,
-0.00027871131896972656,
-0.03778076171875,
0.034332275390625,
-0.01155853271484375,
-0.01213836669921875,
0.050445556640625,
0.048736572265625,
0.043243408203125,
-0.0165252685546875,
0.049407958984375,
0.037109375,
0.01361846923828125,
-0.02667236328125,
0.0614013671875,
0.006069183349609375,
-0.0302734375,
-0.0005021095275878906,
0.00098419189453125,
-0.03631591796875,
0.0213470458984375,
-0.044158935546875,
0.042755126953125,
-0.07745361328125,
-0.007686614990234375,
0.00711822509765625,
0.005069732666015625,
-0.046478271484375,
0.029937744140625,
0.01739501953125,
0.1024169921875,
-0.067626953125,
0.055206298828125,
0.059722900390625,
-0.0142669677734375,
-0.029541015625,
-0.0213165283203125,
0.0085601806640625,
-0.0250244140625,
0.0195465087890625,
0.0165252685546875,
-0.036895751953125,
0.00027179718017578125,
-0.06268310546875,
-0.06939697265625,
0.0836181640625,
0.0191802978515625,
-0.01885986328125,
0.027099609375,
-0.040313720703125,
0.05181884765625,
-0.05126953125,
-0.00701141357421875,
0.008758544921875,
0.025360107421875,
0.06787109375,
-0.01342010498046875,
-0.037750244140625,
-0.063720703125,
0.0305633544921875,
-0.006290435791015625,
-0.0689697265625,
0.043426513671875,
-0.0091705322265625,
-0.0169677734375,
0.045623779296875,
0.042572021484375,
0.0155029296875,
0.0211639404296875,
0.041595458984375,
0.053680419921875,
0.0266876220703125,
0.0024871826171875,
0.09979248046875,
0.0058441162109375,
0.03558349609375,
0.06817626953125,
-0.033935546875,
0.046417236328125,
0.05029296875,
0.0120697021484375,
0.046417236328125,
0.058837890625,
-0.0179901123046875,
0.058746337890625,
-0.01227569580078125,
-0.038116455078125,
-0.015167236328125,
-0.01100921630859375,
-0.0174560546875,
0.017791748046875,
0.0020160675048828125,
0.00467681884765625,
-0.0229644775390625,
-0.0206146240234375,
-0.034210205078125,
-0.00875091552734375,
-0.006542205810546875,
0.031402587890625,
-0.005596160888671875,
-0.0235595703125,
0.0096588134765625,
-0.0160064697265625,
0.002197265625,
0.00127410888671875,
0.003513336181640625,
-0.0034847259521484375,
0.003955841064453125,
-0.034820556640625,
-0.043487548828125,
0.0175628662109375,
-0.032806396484375,
-0.0101776123046875,
-0.00792694091796875,
0.029937744140625,
-0.001270294189453125,
-0.09686279296875,
0.037628173828125,
0.0018291473388671875,
-0.0175933837890625,
0.00701904296875,
-0.05474853515625,
0.014068603515625,
-0.0037994384765625,
-0.0257568359375,
-0.01641845703125,
0.002544403076171875,
0.01393890380859375,
0.0207672119140625,
0.04296875,
0.0183868408203125,
0.004520416259765625,
0.0005326271057128906,
0.051483154296875,
-0.06524658203125,
-0.05828857421875,
-0.021514892578125,
0.039459228515625,
-0.06146240234375,
-0.03515625,
0.0670166015625,
0.079833984375,
0.034942626953125,
-0.035003662109375,
0.046173095703125,
0.01171112060546875,
0.02276611328125,
-0.006744384765625,
0.00821685791015625,
-0.07659912109375,
-0.00873565673828125,
-0.03448486328125,
-0.09454345703125,
-0.060546875,
0.04681396484375,
0.0019855499267578125,
-0.004329681396484375,
0.0142669677734375,
0.05035400390625,
-0.0260467529296875,
0.029205322265625,
0.023956298828125,
0.0180511474609375,
0.022308349609375,
-0.0026569366455078125,
0.0726318359375,
-0.04498291015625,
-0.00592803955078125,
-0.044830322265625,
-0.0305938720703125,
-0.033935546875,
-0.06304931640625,
-0.040863037109375,
-0.02655029296875,
-0.0283050537109375,
-0.0127105712890625,
-0.018646240234375,
0.03765869140625,
0.0634765625,
-0.038360595703125,
-0.016754150390625,
-0.03631591796875,
-0.0017690658569335938,
0.0111236572265625,
-0.01678466796875,
-0.0249176025390625,
0.03021240234375,
-0.0379638671875,
0.0257568359375,
0.04327392578125,
0.03936767578125,
-0.01230621337890625,
0.039642333984375,
0.01160430908203125,
0.023040771484375,
0.03814697265625,
0.00814056396484375,
-0.05548095703125,
-0.0305023193359375,
0.0175628662109375,
-0.003543853759765625,
0.006519317626953125,
0.0618896484375,
-0.006237030029296875,
-0.00922393798828125,
0.033477783203125,
-0.029205322265625,
0.05377197265625,
-0.00926971435546875,
0.0205230712890625,
-0.0183563232421875,
0.05126953125,
0.003692626953125,
0.041473388671875,
0.018280029296875,
-0.0306243896484375,
0.034698486328125,
0.0264892578125,
-0.03717041015625,
-0.0526123046875,
0.01360321044921875,
-0.11822509765625,
-0.0178070068359375,
0.052093505859375,
0.04022216796875,
-0.043426513671875,
0.025299072265625,
-0.036834716796875,
-0.04351806640625,
-0.0162353515625,
0.031036376953125,
0.06671142578125,
-0.01360321044921875,
-0.0301361083984375,
-0.0634765625,
-0.00249481201171875,
0.0006313323974609375,
-0.044158935546875,
-0.04620361328125,
0.059478759765625,
0.047943115234375,
0.022308349609375,
0.039306640625,
-0.046173095703125,
0.0177154541015625,
0.0200653076171875,
0.033905029296875,
0.028472900390625,
-0.048187255859375,
0.00934600830078125,
0.02001953125,
0.0124053955078125,
-0.022613525390625
]
] |
heegyu/LIMA-13b-hf | 2023-08-01T12:49:55.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"license:other",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | heegyu | null | null | heegyu/LIMA-13b-hf | 1 | 7,562 | transformers | 2023-08-01T12:29:28 | ---
license: other
---
LLaMA-13B converted to work with Transformers/HuggingFace. This is under a special license, please see the LICENSE file for details.
--
license: other
---
# LLaMA Model Card
## Model details
**Organization developing the model**
The FAIR team of Meta AI.
**Model date**
LLaMA was trained between December. 2022 and Feb. 2023.
**Model version**
This is version 1 of the model.
**Model type**
LLaMA is an auto-regressive language model, based on the transformer architecture. The model comes in different sizes: 7B, 13B, 33B and 65B parameters.
**Paper or resources for more information**
More information can be found in the paper “LLaMA, Open and Efficient Foundation Language Models”, available at https://research.facebook.com/publications/llama-open-and-efficient-foundation-language-models/.
**Citations details**
https://research.facebook.com/publications/llama-open-and-efficient-foundation-language-models/
**License**
Non-commercial bespoke license
**Where to send questions or comments about the model**
Questions and comments about LLaMA can be sent via the [GitHub repository](https://github.com/facebookresearch/llama) of the project , by opening an issue.
## Intended use
**Primary intended uses**
The primary use of LLaMA is research on large language models, including:
exploring potential applications such as question answering, natural language understanding or reading comprehension,
understanding capabilities and limitations of current language models, and developing techniques to improve those,
evaluating and mitigating biases, risks, toxic and harmful content generations, hallucinations.
**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**
LLaMA 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.
## Factors
**Relevant factors**
One of the most relevant factors for which model performance may vary is which language is used. Although we included 20 languages in the training data, most of our dataset is made of English text, and we thus expect the model to perform better for English than other languages. Relatedly, it has been shown in previous studies that performance might vary for different dialects, and we expect that it will be the case for our model.
**Evaluation factors**
As our model is trained on data from the Web, we expect that it reflects biases from this source. We thus evaluated on RAI datasets to measure biases exhibited by the model for gender, religion, race, sexual orientation, age, nationality, disability, physical appearance and socio-economic status. We also measure the toxicity of model generations, depending on the toxicity of the context used to prompt the model.
## Metrics
**Model performance measures**
We use the following measure to evaluate the model:
- Accuracy for common sense reasoning, reading comprehension, natural language understanding (MMLU), BIG-bench hard, WinoGender and CrowS-Pairs,
- Exact match for question answering,
- The toxicity score from Perspective API on RealToxicityPrompts.
**Decision thresholds**
Not applicable.
**Approaches to uncertainty and variability**
Due to the high computational requirements of training LLMs, we trained only one model of each size, and thus could not evaluate variability of pre-training.
## Evaluation datasets
The model was evaluated on the following benchmarks: BoolQ, PIQA, SIQA, HellaSwag, WinoGrande, ARC, OpenBookQA, NaturalQuestions, TriviaQA, RACE, MMLU, BIG-bench hard, GSM8k, RealToxicityPrompts, WinoGender, CrowS-Pairs.
## Training dataset
The model was trained using the following source of data: CCNet [67%], C4 [15%], GitHub [4.5%], Wikipedia [4.5%], Books [4.5%], ArXiv [2.5%], Stack Exchange[2%]. The Wikipedia and Books domains include data in the following languages: bg, ca, cs, da, de, en, es, fr, hr, hu, it, nl, pl, pt, ro, ru, sl, sr, sv, uk. See the paper for more details about the training set and corresponding preprocessing.
## Quantitative analysis
Hyperparameters for the model architecture
<table>
<thead>
<tr>
<th >LLaMA</th> <th colspan=6>Model hyper parameters </th>
</tr>
<tr>
<th>Number of parameters</th><th>dimension</th><th>n heads</th><th>n layers</th><th>Learn rate</th><th>Batch size</th><th>n tokens</th>
</tr>
</thead>
<tbody>
<tr>
<th>7B</th> <th>4096</th> <th>32</th> <th>32</th> <th>3.0E-04</th><th>4M</th><th>1T
</tr>
<tr>
<th>13B</th><th>5120</th><th>40</th><th>40</th><th>3.0E-04</th><th>4M</th><th>1T
</tr>
<tr>
<th>33B</th><th>6656</th><th>52</th><th>60</th><th>1.5.E-04</th><th>4M</th><th>1.4T
</tr>
<tr>
<th>65B</th><th>8192</th><th>64</th><th>80</th><th>1.5.E-04</th><th>4M</th><th>1.4T
</tr>
</tbody>
</table>
*Table 1 - Summary of LLama Model Hyperparameters*
We present our results on eight standard common sense reasoning benchmarks in the table below.
<table>
<thead>
<tr>
<th>LLaMA</th> <th colspan=9>Reasoning tasks </th>
</tr>
<tr>
<th>Number of parameters</th> <th>BoolQ</th><th>PIQA</th><th>SIQA</th><th>HellaSwag</th><th>WinoGrande</th><th>ARC-e</th><th>ARC-c</th><th>OBQA</th><th>COPA</th>
</tr>
</thead>
<tbody>
<tr>
<th>7B</th><th>76.5</th><th>79.8</th><th>48.9</th><th>76.1</th><th>70.1</th><th>76.7</th><th>47.6</th><th>57.2</th><th>93
</th>
<tr><th>13B</th><th>78.1</th><th>80.1</th><th>50.4</th><th>79.2</th><th>73</th><th>78.1</th><th>52.7</th><th>56.4</th><th>94
</th>
<tr><th>33B</th><th>83.1</th><th>82.3</th><th>50.4</th><th>82.8</th><th>76</th><th>81.4</th><th>57.8</th><th>58.6</th><th>92
</th>
<tr><th>65B</th><th>85.3</th><th>82.8</th><th>52.3</th><th>84.2</th><th>77</th><th>81.5</th><th>56</th><th>60.2</th><th>94</th></tr>
</tbody>
</table>
*Table 2 - Summary of LLama Model Performance on Reasoning tasks*
We present our results on bias in the table below. Note that lower value is better indicating lower bias.
| No | Category | FAIR LLM |
| --- | -------------------- | -------- |
| 1 | Gender | 70.6 |
| 2 | Religion | 79 |
| 3 | Race/Color | 57 |
| 4 | Sexual orientation | 81 |
| 5 | Age | 70.1 |
| 6 | Nationality | 64.2 |
| 7 | Disability | 66.7 |
| 8 | Physical appearance | 77.8 |
| 9 | Socioeconomic status | 71.5 |
| | LLaMA Average | 66.6 |
*Table 3 - Summary bias of our model output*
## 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.
**Mitigations**
We filtered the data from the Web based on its proximity to Wikipedia text and references. For this, we used a Kneser-Ney language model and a fastText linear classifier.
**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**
LLaMA 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.
| 8,309 | [
[
-0.0298004150390625,
-0.05517578125,
0.032928466796875,
0.021942138671875,
-0.0169525146484375,
-0.0185394287109375,
0.0008168220520019531,
-0.049102783203125,
0.005413055419921875,
0.0311279296875,
-0.03643798828125,
-0.042327880859375,
-0.054229736328125,
0.016082763671875,
-0.032928466796875,
0.06304931640625,
0.0098114013671875,
-0.0052490234375,
-0.01202392578125,
-0.015960693359375,
-0.0199737548828125,
-0.0472412109375,
-0.03729248046875,
-0.0211944580078125,
0.03594970703125,
0.0178680419921875,
0.05389404296875,
0.052459716796875,
0.03314208984375,
0.025146484375,
-0.0305328369140625,
0.01503753662109375,
-0.0198516845703125,
-0.0100250244140625,
-0.01495361328125,
-0.028717041015625,
-0.03948974609375,
0.00732421875,
0.0215606689453125,
0.04339599609375,
-0.0180511474609375,
0.04083251953125,
0.0035343170166015625,
0.048187255859375,
-0.05584716796875,
0.00421142578125,
-0.04345703125,
0.0005102157592773438,
-0.0068817138671875,
0.006103515625,
-0.0140533447265625,
-0.025604248046875,
0.0021953582763671875,
-0.0548095703125,
-0.01152801513671875,
0.012786865234375,
0.08984375,
0.0484619140625,
-0.039703369140625,
-0.0014810562133789062,
-0.0246124267578125,
0.082275390625,
-0.07452392578125,
0.034820556640625,
0.02593994140625,
0.0174713134765625,
-0.0109405517578125,
-0.040283203125,
-0.060394287109375,
-0.021484375,
0.00678253173828125,
0.0166473388671875,
-0.0284881591796875,
-0.01213836669921875,
0.033050537109375,
0.01293182373046875,
-0.0313720703125,
0.0180511474609375,
-0.0208740234375,
-0.0164337158203125,
0.06500244140625,
0.027984619140625,
0.00928497314453125,
-0.021209716796875,
-0.036346435546875,
-0.00466156005859375,
-0.042755126953125,
0.012847900390625,
0.027587890625,
0.01404571533203125,
-0.0154266357421875,
0.048492431640625,
-0.0328369140625,
0.04925537109375,
-0.002185821533203125,
-0.03765869140625,
0.03594970703125,
-0.03131103515625,
-0.025909423828125,
-0.007518768310546875,
0.06414794921875,
0.037139892578125,
0.0269317626953125,
0.0101470947265625,
-0.005809783935546875,
0.0185394287109375,
0.0052947998046875,
-0.0567626953125,
0.0152130126953125,
0.0122528076171875,
-0.033477783203125,
-0.0220184326171875,
-0.0289764404296875,
-0.05291748046875,
-0.02362060546875,
-0.032196044921875,
0.0046844482421875,
-0.0220794677734375,
-0.0242767333984375,
0.0214996337890625,
0.00868988037109375,
0.027069091796875,
0.0272674560546875,
-0.045928955078125,
0.023834228515625,
0.034271240234375,
0.049072265625,
-0.0247650146484375,
-0.03955078125,
0.0008130073547363281,
0.00948333740234375,
-0.0168609619140625,
0.0657958984375,
-0.027374267578125,
-0.02130126953125,
-0.0206298828125,
-0.00121307373046875,
-0.007678985595703125,
-0.04296875,
0.05267333984375,
-0.032470703125,
0.01261138916015625,
-0.035736083984375,
-0.0386962890625,
-0.0264892578125,
0.04022216796875,
-0.04925537109375,
0.1009521484375,
-0.014312744140625,
-0.061309814453125,
0.022186279296875,
-0.052276611328125,
-0.004901885986328125,
-0.0254058837890625,
-0.0014314651489257812,
-0.0181121826171875,
-0.026580810546875,
0.0255279541015625,
0.0401611328125,
-0.0509033203125,
0.039306640625,
-0.0190582275390625,
-0.039947509765625,
0.019195556640625,
-0.04278564453125,
0.0853271484375,
0.0177764892578125,
-0.042816162109375,
-0.00016200542449951172,
-0.0765380859375,
-0.007785797119140625,
0.039093017578125,
-0.038909912109375,
-0.01404571533203125,
-0.005321502685546875,
-0.001811981201171875,
0.0284881591796875,
0.0202789306640625,
-0.032867431640625,
0.004299163818359375,
-0.032379150390625,
0.025787353515625,
0.0545654296875,
0.00901031494140625,
0.0168609619140625,
-0.033966064453125,
0.03662109375,
0.0201263427734375,
0.01873779296875,
0.0213775634765625,
-0.04949951171875,
-0.07012939453125,
-0.01459503173828125,
0.0164337158203125,
0.0576171875,
-0.0182647705078125,
0.0384521484375,
-0.01800537109375,
-0.051300048828125,
-0.034271240234375,
0.01904296875,
0.04278564453125,
0.05902099609375,
0.03564453125,
-0.002960205078125,
-0.039794921875,
-0.07745361328125,
-0.01093292236328125,
-0.0294342041015625,
0.0151519775390625,
0.033843994140625,
0.048919677734375,
-0.032196044921875,
0.057342529296875,
-0.038421630859375,
-0.01123809814453125,
-0.0164031982421875,
-0.021514892578125,
0.0209808349609375,
0.021392822265625,
0.044586181640625,
-0.05047607421875,
-0.01910400390625,
-0.0077972412109375,
-0.06988525390625,
-0.01126861572265625,
0.0158843994140625,
-0.014007568359375,
0.023590087890625,
0.0276641845703125,
-0.056182861328125,
0.037384033203125,
0.038909912109375,
-0.0188446044921875,
0.058135986328125,
0.0208587646484375,
-0.0034351348876953125,
-0.08270263671875,
0.01580810546875,
0.0180511474609375,
0.01690673828125,
-0.033111572265625,
0.00839996337890625,
-0.01020050048828125,
-0.0008544921875,
-0.057220458984375,
0.0616455078125,
-0.01123809814453125,
0.0090789794921875,
-0.0102386474609375,
0.004383087158203125,
0.00875091552734375,
0.0538330078125,
0.003910064697265625,
0.072021484375,
0.0305633544921875,
-0.05548095703125,
0.0135345458984375,
0.01221466064453125,
-0.02801513671875,
0.0256195068359375,
-0.0673828125,
-0.0008816719055175781,
0.011444091796875,
0.0234832763671875,
-0.0576171875,
-0.00817108154296875,
0.033172607421875,
-0.0311431884765625,
-0.0145111083984375,
0.00821685791015625,
-0.05029296875,
-0.0264434814453125,
-0.01114654541015625,
0.022186279296875,
0.034576416015625,
-0.033843994140625,
0.0360107421875,
0.0263214111328125,
0.00716400146484375,
-0.072021484375,
-0.06158447265625,
-0.004756927490234375,
-0.027130126953125,
-0.0615234375,
0.01454925537109375,
-0.01070404052734375,
-0.034454345703125,
-0.005344390869140625,
-0.004825592041015625,
-0.00518035888671875,
0.0208892822265625,
0.0171661376953125,
0.018829345703125,
-0.0110321044921875,
0.007694244384765625,
-0.0005321502685546875,
-0.01221466064453125,
0.00884246826171875,
0.01007080078125,
0.04058837890625,
-0.0216827392578125,
-0.038177490234375,
-0.0465087890625,
0.00951385498046875,
0.033416748046875,
-0.0158843994140625,
0.0540771484375,
0.033905029296875,
-0.02740478515625,
0.0015153884887695312,
-0.05303955078125,
-0.004085540771484375,
-0.03759765625,
0.030303955078125,
-0.0203399658203125,
-0.060272216796875,
0.047637939453125,
-0.0091552734375,
-0.005519866943359375,
0.045806884765625,
0.05108642578125,
0.005096435546875,
0.07696533203125,
0.035614013671875,
-0.022705078125,
0.0229034423828125,
-0.0294036865234375,
0.01511383056640625,
-0.062744140625,
-0.0308380126953125,
-0.038482666015625,
-0.017181396484375,
-0.0338134765625,
-0.0287933349609375,
0.014739990234375,
-0.0009036064147949219,
-0.050445556640625,
0.014556884765625,
-0.048187255859375,
0.0262451171875,
0.049896240234375,
0.01251983642578125,
0.0267333984375,
-0.004848480224609375,
-0.026153564453125,
-0.001338958740234375,
-0.04498291015625,
-0.055999755859375,
0.09033203125,
0.032684326171875,
0.045806884765625,
0.004886627197265625,
0.045196533203125,
0.0021915435791015625,
-0.00130462646484375,
-0.05517578125,
0.06427001953125,
0.0180511474609375,
-0.06280517578125,
-0.0171966552734375,
-0.01971435546875,
-0.06842041015625,
0.034881591796875,
-0.0058135986328125,
-0.06097412109375,
0.030364990234375,
-0.0012035369873046875,
-0.0193023681640625,
0.0258026123046875,
-0.0570068359375,
0.047393798828125,
-0.0305633544921875,
-0.0290985107421875,
-0.01092529296875,
-0.044708251953125,
0.042022705078125,
0.0030689239501953125,
0.021759033203125,
-0.0229949951171875,
-0.0176544189453125,
0.06939697265625,
-0.0299835205078125,
0.07281494140625,
-0.006183624267578125,
0.004581451416015625,
0.031158447265625,
-0.0068511962890625,
0.03997802734375,
-0.0011167526245117188,
-0.0281524658203125,
0.035430908203125,
-0.0062408447265625,
-0.033966064453125,
-0.033966064453125,
0.048492431640625,
-0.080322265625,
-0.06903076171875,
-0.050323486328125,
-0.046173095703125,
-0.01438140869140625,
0.0167388916015625,
0.01113128662109375,
-0.0114898681640625,
0.01192474365234375,
0.00530242919921875,
0.05377197265625,
-0.02911376953125,
0.0227813720703125,
0.0290679931640625,
-0.001514434814453125,
-0.0157318115234375,
0.059295654296875,
0.0130767822265625,
0.0135040283203125,
0.002849578857421875,
0.01666259765625,
-0.0224761962890625,
-0.0416259765625,
-0.029296875,
0.02752685546875,
-0.0572509765625,
-0.0184173583984375,
-0.059112548828125,
-0.0253753662109375,
-0.029296875,
-0.0007791519165039062,
-0.02862548828125,
-0.027740478515625,
-0.036224365234375,
-0.0220184326171875,
0.038421630859375,
0.036163330078125,
0.0150299072265625,
0.0271148681640625,
-0.031280517578125,
0.00890350341796875,
0.0184173583984375,
0.00852203369140625,
0.0180511474609375,
-0.043792724609375,
-0.006744384765625,
0.002597808837890625,
-0.039794921875,
-0.059661865234375,
0.028289794921875,
-0.01389312744140625,
0.063232421875,
0.02850341796875,
0.0003833770751953125,
0.0548095703125,
-0.01082611083984375,
0.08074951171875,
0.01546478271484375,
-0.06256103515625,
0.043182373046875,
-0.023681640625,
0.0214080810546875,
0.043243408203125,
0.04400634765625,
-0.021881103515625,
-0.0099639892578125,
-0.049468994140625,
-0.0634765625,
0.0404052734375,
0.0074462890625,
-0.003704071044921875,
-0.00046944618225097656,
0.01220703125,
-0.00890350341796875,
0.01474761962890625,
-0.076171875,
-0.034027099609375,
-0.0133209228515625,
-0.006500244140625,
0.010528564453125,
-0.019683837890625,
-0.01873779296875,
-0.043548583984375,
0.053192138671875,
-0.0019779205322265625,
0.020782470703125,
0.0020999908447265625,
-0.01485443115234375,
0.009124755859375,
0.016693115234375,
0.035003662109375,
0.058868408203125,
-0.00982666015625,
-0.00678253173828125,
0.036407470703125,
-0.04443359375,
0.0186614990234375,
0.01004791259765625,
-0.0225982666015625,
-0.02166748046875,
0.0272979736328125,
0.05560302734375,
0.01561737060546875,
-0.06243896484375,
0.033477783203125,
0.0108642578125,
-0.01934814453125,
-0.019683837890625,
0.01218414306640625,
0.01561737060546875,
0.0301055908203125,
0.0176544189453125,
-0.0003693103790283203,
0.016204833984375,
-0.02081298828125,
-0.0129241943359375,
0.0161285400390625,
-0.0054473876953125,
-0.0084075927734375,
0.057647705078125,
0.00670623779296875,
-0.01050567626953125,
0.03662109375,
-0.0070648193359375,
-0.0338134765625,
0.055755615234375,
0.0401611328125,
0.038116455078125,
-0.00830078125,
0.009307861328125,
0.03955078125,
0.0250701904296875,
-0.006134033203125,
0.0257415771484375,
0.00482940673828125,
-0.047698974609375,
-0.0233612060546875,
-0.06085205078125,
-0.034271240234375,
0.0112762451171875,
-0.043182373046875,
0.015472412109375,
-0.03765869140625,
-0.0168914794921875,
-0.008026123046875,
0.02423095703125,
-0.06231689453125,
0.0333251953125,
0.00698089599609375,
0.0855712890625,
-0.06451416015625,
0.058868408203125,
0.054351806640625,
-0.051788330078125,
-0.07720947265625,
-0.007770538330078125,
0.014801025390625,
-0.07415771484375,
0.05419921875,
0.004215240478515625,
-0.0013170242309570312,
-0.00742340087890625,
-0.052490234375,
-0.095703125,
0.10687255859375,
0.03106689453125,
-0.0367431640625,
-0.0011034011840820312,
0.0276641845703125,
0.043121337890625,
-0.0161895751953125,
0.0216827392578125,
0.044525146484375,
0.043365478515625,
0.01123809814453125,
-0.07171630859375,
0.0219879150390625,
-0.03955078125,
-0.0010404586791992188,
-0.015716552734375,
-0.07232666015625,
0.07110595703125,
-0.006137847900390625,
0.005218505859375,
-0.005008697509765625,
0.041351318359375,
0.071533203125,
0.036834716796875,
0.035858154296875,
0.0701904296875,
0.06695556640625,
0.00821685791015625,
0.08673095703125,
-0.0208587646484375,
0.0080718994140625,
0.0675048828125,
-0.0220947265625,
0.06866455078125,
0.0372314453125,
-0.038482666015625,
0.039398193359375,
0.0655517578125,
0.0032520294189453125,
0.026397705078125,
0.0109405517578125,
-0.0007023811340332031,
-0.0181121826171875,
-0.0256195068359375,
-0.037384033203125,
0.0281982421875,
0.0215606689453125,
-0.02581787109375,
-0.00994873046875,
-0.009368896484375,
0.016448974609375,
0.00081634521484375,
-0.0122528076171875,
0.0382080078125,
0.0098419189453125,
-0.038726806640625,
0.062347412109375,
-0.0038356781005859375,
0.064208984375,
-0.039703369140625,
0.006011962890625,
-0.0199737548828125,
0.001155853271484375,
-0.0220947265625,
-0.046783447265625,
0.0110015869140625,
0.007678985595703125,
-0.01361846923828125,
0.007022857666015625,
0.05462646484375,
0.0180511474609375,
-0.05426025390625,
0.04437255859375,
0.04510498046875,
0.01561737060546875,
-0.0036449432373046875,
-0.07958984375,
0.0220184326171875,
-0.0084228515625,
-0.05145263671875,
0.0284576416015625,
0.0289764404296875,
-0.025421142578125,
0.0806884765625,
0.05853271484375,
0.00547027587890625,
0.024810791015625,
0.0036373138427734375,
0.08270263671875,
-0.0550537109375,
-0.0169677734375,
-0.0540771484375,
0.0406494140625,
-0.002300262451171875,
-0.0423583984375,
0.06414794921875,
0.035797119140625,
0.0635986328125,
0.022003173828125,
0.057525634765625,
0.00722503662109375,
0.038299560546875,
-0.0234222412109375,
0.047515869140625,
-0.055389404296875,
0.0229644775390625,
-0.0229034423828125,
-0.0753173828125,
-0.0211944580078125,
0.0439453125,
-0.038299560546875,
0.01099395751953125,
0.04180908203125,
0.06658935546875,
-0.0036983489990234375,
-0.00327301025390625,
0.0141448974609375,
0.0209197998046875,
0.011138916015625,
0.041473388671875,
0.060272216796875,
-0.035308837890625,
0.046661376953125,
-0.033538818359375,
-0.00876617431640625,
-0.012420654296875,
-0.05413818359375,
-0.06488037109375,
-0.035369873046875,
-0.01971435546875,
-0.022491455078125,
0.0018587112426757812,
0.06475830078125,
0.050140380859375,
-0.0498046875,
-0.0282135009765625,
-0.0005517005920410156,
0.011993408203125,
-0.01702880859375,
-0.01543426513671875,
0.030181884765625,
-0.004611968994140625,
-0.05206298828125,
0.01433563232421875,
-0.004604339599609375,
-0.0051116943359375,
-0.0246734619140625,
-0.0214691162109375,
-0.049346923828125,
0.01453399658203125,
0.043060302734375,
0.01100921630859375,
-0.07940673828125,
-0.01074981689453125,
0.0178680419921875,
-0.0166778564453125,
-0.0016498565673828125,
0.00203704833984375,
-0.057830810546875,
0.0021839141845703125,
0.0189361572265625,
0.058502197265625,
0.037506103515625,
-0.00318145751953125,
0.00982666015625,
-0.0298309326171875,
0.0179290771484375,
0.01812744140625,
0.0260162353515625,
0.01511383056640625,
-0.03857421875,
0.062042236328125,
0.0296783447265625,
-0.048187255859375,
-0.07574462890625,
-0.0014677047729492188,
-0.0875244140625,
-0.0173797607421875,
0.107421875,
-0.0020503997802734375,
-0.0369873046875,
0.00044608116149902344,
-0.0183258056640625,
0.0257415771484375,
-0.034515380859375,
0.04962158203125,
0.05462646484375,
-0.017181396484375,
-0.0128631591796875,
-0.058563232421875,
0.01561737060546875,
0.0382080078125,
-0.0618896484375,
-0.0188446044921875,
0.033050537109375,
0.027435302734375,
0.012451171875,
0.038330078125,
-0.004261016845703125,
0.021697998046875,
-0.00286865234375,
0.01078033447265625,
-0.01511383056640625,
-0.01495361328125,
-0.0145111083984375,
-0.003231048583984375,
0.0059051513671875,
-0.006038665771484375
]
] |
heegyu/LIMA2-7b-hf | 2023-08-04T07:53:14.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"facebook",
"meta",
"llama-2",
"en",
"arxiv:2307.09288",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | heegyu | null | null | heegyu/LIMA2-7b-hf | 1 | 7,562 | transformers | 2023-08-04T06:55:07 | ---
extra_gated_heading: Access Llama 2 on Hugging Face
extra_gated_description: >-
This is a form to enable access to Llama 2 on Hugging Face after you have been
granted access from Meta. Please visit the [Meta website](https://ai.meta.com/resources/models-and-libraries/llama-downloads) and accept our
license terms and acceptable use policy before submitting this form. Requests
will be processed in 1-2 days.
extra_gated_prompt: "**Your Hugging Face account email address MUST match the email you provide on the Meta website, or your request will not be approved.**"
extra_gated_button_content: Submit
extra_gated_fields:
I agree to share my name, email address and username with Meta and confirm that I have already been granted download access on the Meta website: checkbox
language:
- en
pipeline_tag: text-generation
inference: false
tags:
- facebook
- meta
- pytorch
- llama
- llama-2
---
finetuned Llama-2-7b-hf model with 64bits/lima_vicuna_format data (10 epoch)
### prompt
```
### Human:
Hi, how are you?
### Assistant:
```
# **Llama 2**
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 7B pretrained model. Links to other models can be found in the index at the bottom.
## Model Details
*Note: Use of this model is governed by the Meta license. 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.*
Meta developed and publicly released the Llama 2 family of large language models (LLMs), a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama-2-Chat, are optimized for dialogue use cases. Llama-2-Chat models outperform open-source chat models on most benchmarks we tested, and in our human evaluations for helpfulness and safety, are on par with some popular closed-source models like ChatGPT and PaLM.
**Model Developers** Meta
**Variations** Llama 2 comes in a range of parameter sizes — 7B, 13B, and 70B — as well as pretrained and fine-tuned variations.
**Input** Models input text only.
**Output** Models generate text only.
**Model Architecture** Llama 2 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.
||Training Data|Params|Content Length|GQA|Tokens|LR|
|---|---|---|---|---|---|---|
|Llama 2|*A new mix of publicly available online data*|7B|4k|✗|2.0T|3.0 x 10<sup>-4</sup>|
|Llama 2|*A new mix of publicly available online data*|13B|4k|✗|2.0T|3.0 x 10<sup>-4</sup>|
|Llama 2|*A new mix of publicly available online data*|70B|4k|✔|2.0T|1.5 x 10<sup>-4</sup>|
*Llama 2 family of models.* Token counts refer to pretraining data only. All models are trained with a global batch-size of 4M tokens. Bigger models - 70B -- use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Dates** Llama 2 was trained between January 2023 and July 2023.
**Status** This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.
**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** ["Llama-2: Open Foundation and Fine-tuned Chat Models"](arxiv.org/abs/2307.09288)
## Intended Use
**Intended Use Cases** Llama 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
To get the expected features and performance for the chat versions, a specific formatting needs to be followed, including the `INST` and `<<SYS>>` tags, `BOS` and `EOS` tokens, and the whitespaces and breaklines in between (we recommend calling `strip()` on inputs to avoid double-spaces). See our reference code in github for details: [`chat_completion`](https://github.com/facebookresearch/llama/blob/main/llama/generation.py#L212).
**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 Llama 2.
## Hardware and Software
**Training Factors** We used custom training libraries, Meta's Research Super Cluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.
**Carbon Footprint** Pretraining utilized a cumulative 3.3M GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 539 tCO2eq, 100% of which were offset by Meta’s sustainability program.
||Time (GPU hours)|Power Consumption (W)|Carbon Emitted(tCO<sub>2</sub>eq)|
|---|---|---|---|
|Llama 2 7B|184320|400|31.22|
|Llama 2 13B|368640|400|62.44|
|Llama 2 70B|1720320|400|291.42|
|Total|3311616||539.00|
**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** Llama 2 was pretrained on 2 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over one million new human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.
**Data Freshness** The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023.
## Evaluation Results
In this section, we report the results for the Llama 1 and Llama 2 models on standard academic benchmarks.For all the evaluations, we use our internal evaluations library.
|Model|Size|Code|Commonsense Reasoning|World Knowledge|Reading Comprehension|Math|MMLU|BBH|AGI Eval|
|---|---|---|---|---|---|---|---|---|---|
|Llama 1|7B|14.1|60.8|46.2|58.5|6.95|35.1|30.3|23.9|
|Llama 1|13B|18.9|66.1|52.6|62.3|10.9|46.9|37.0|33.9|
|Llama 1|33B|26.0|70.0|58.4|67.6|21.4|57.8|39.8|41.7|
|Llama 1|65B|30.7|70.7|60.5|68.6|30.8|63.4|43.5|47.6|
|Llama 2|7B|16.8|63.9|48.9|61.3|14.6|45.3|32.6|29.3|
|Llama 2|13B|24.5|66.9|55.4|65.8|28.7|54.8|39.4|39.1|
|Llama 2|70B|**37.5**|**71.9**|**63.6**|**69.4**|**35.2**|**68.9**|**51.2**|**54.2**|
**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
Llama 2 is 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’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, 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/](https://ai.meta.com/llama/responsible-use-guide)
## Reporting Issues
Please report any software “bug,” or other problems with the models through one of the following means:
- Reporting issues with the model: [github.com/facebookresearch/llama](http://github.com/facebookresearch/llama)
- Reporting problematic content generated by the model: [developers.facebook.com/llama_output_feedback](http://developers.facebook.com/llama_output_feedback)
- Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)
## Llama Model Index
|Model|Llama2|Llama2-hf|Llama2-chat|Llama2-chat-hf|
|---|---|---|---|---|
|7B| [Link](https://huggingface.co/llamaste/Llama-2-7b) | [Link](https://huggingface.co/llamaste/Llama-2-7b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat-hf)|
|13B| [Link](https://huggingface.co/llamaste/Llama-2-13b) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-13b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf)|
|70B| [Link](https://huggingface.co/llamaste/Llama-2-70b) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-70b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf)| | 10,449 | [
[
-0.0169830322265625,
-0.054229736328125,
0.0285186767578125,
0.01374053955078125,
-0.02996826171875,
0.0164337158203125,
-0.005336761474609375,
-0.056732177734375,
0.0048370361328125,
0.0224609375,
-0.052764892578125,
-0.042236328125,
-0.04998779296875,
0.00727081298828125,
-0.01436614990234375,
0.07818603515625,
0.0023593902587890625,
-0.021026611328125,
-0.00923919677734375,
0.007659912109375,
-0.037322998046875,
-0.0305328369140625,
-0.03961181640625,
-0.032928466796875,
0.0284271240234375,
0.03863525390625,
0.045013427734375,
0.050689697265625,
0.041534423828125,
0.0177459716796875,
-0.020355224609375,
0.017120361328125,
-0.05523681640625,
-0.0219573974609375,
0.01129913330078125,
-0.036590576171875,
-0.049560546875,
0.01009368896484375,
0.02752685546875,
0.01548004150390625,
-0.021942138671875,
0.039794921875,
0.0067901611328125,
0.03558349609375,
-0.041259765625,
0.0146026611328125,
-0.05474853515625,
0.0014019012451171875,
-0.0172882080078125,
-0.005931854248046875,
-0.01462554931640625,
-0.0199737548828125,
-0.01447296142578125,
-0.06280517578125,
-0.0083160400390625,
0.006595611572265625,
0.07861328125,
0.048248291015625,
-0.032012939453125,
-0.0090484619140625,
-0.02215576171875,
0.06829833984375,
-0.066650390625,
0.0057373046875,
0.04449462890625,
0.0204315185546875,
-0.0198211669921875,
-0.05584716796875,
-0.049530029296875,
-0.01168060302734375,
0.005764007568359375,
0.0247650146484375,
-0.0297393798828125,
0.0005431175231933594,
0.0115966796875,
0.0276947021484375,
-0.042999267578125,
0.0428466796875,
-0.0404052734375,
-0.01432037353515625,
0.07745361328125,
0.018341064453125,
-0.002285003662109375,
-0.005802154541015625,
-0.0362548828125,
-0.0194549560546875,
-0.0599365234375,
0.01305389404296875,
0.03778076171875,
-0.0015954971313476562,
-0.035675048828125,
0.046905517578125,
-0.032501220703125,
0.022003173828125,
0.0041656494140625,
-0.03887939453125,
0.037139892578125,
-0.034149169921875,
-0.0225067138671875,
-0.0081024169921875,
0.06829833984375,
0.053924560546875,
0.0096893310546875,
0.0088958740234375,
-0.003753662109375,
0.00794219970703125,
-0.002197265625,
-0.062744140625,
-0.004306793212890625,
0.0187835693359375,
-0.02874755859375,
-0.042266845703125,
-0.0214996337890625,
-0.054718017578125,
-0.01410675048828125,
-0.00733184814453125,
0.0184326171875,
-0.00336456298828125,
-0.027923583984375,
0.00952911376953125,
0.00493621826171875,
0.040435791015625,
0.0169219970703125,
-0.0704345703125,
0.0179595947265625,
0.0400390625,
0.058441162109375,
-0.0179901123046875,
-0.0255889892578125,
0.002971649169921875,
-0.00281524658203125,
-0.0239715576171875,
0.06622314453125,
-0.0245513916015625,
-0.0391845703125,
-0.0187835693359375,
-0.001739501953125,
0.0146484375,
-0.03948974609375,
0.033111572265625,
-0.0295867919921875,
0.014404296875,
-0.0246429443359375,
-0.027587890625,
-0.0254364013671875,
0.01538848876953125,
-0.0302581787109375,
0.10821533203125,
0.0092010498046875,
-0.038116455078125,
0.0226287841796875,
-0.050323486328125,
-0.0139923095703125,
-0.01531219482421875,
0.00787353515625,
-0.04034423828125,
-0.0226593017578125,
0.01033782958984375,
0.028533935546875,
-0.050140380859375,
0.037017822265625,
-0.0172576904296875,
-0.03448486328125,
0.003925323486328125,
-0.0352783203125,
0.0638427734375,
0.02252197265625,
-0.034637451171875,
0.00328826904296875,
-0.06427001953125,
0.0027313232421875,
0.035064697265625,
-0.037811279296875,
0.02020263671875,
0.003986358642578125,
-0.00897979736328125,
0.0145721435546875,
0.038787841796875,
-0.02801513671875,
0.0112152099609375,
-0.025634765625,
0.03729248046875,
0.05523681640625,
0.003936767578125,
0.0124969482421875,
-0.038482666015625,
0.037139892578125,
-0.0019350051879882812,
0.0308380126953125,
0.001819610595703125,
-0.054229736328125,
-0.07666015625,
-0.01340484619140625,
-0.00202178955078125,
0.06182861328125,
-0.018035888671875,
0.052490234375,
-0.0012798309326171875,
-0.056732177734375,
-0.033203125,
0.0295867919921875,
0.05035400390625,
0.038482666015625,
0.032623291015625,
-0.0224456787109375,
-0.046600341796875,
-0.07586669921875,
0.0085601806640625,
-0.03375244140625,
-0.00140380859375,
0.02777099609375,
0.0477294921875,
-0.025604248046875,
0.05487060546875,
-0.042022705078125,
-0.01389312744140625,
-0.0204010009765625,
-0.0110015869140625,
0.006542205810546875,
0.0269775390625,
0.04766845703125,
-0.028839111328125,
-0.0157623291015625,
-0.00928497314453125,
-0.066650390625,
-0.007472991943359375,
0.008697509765625,
-0.0157470703125,
0.0188751220703125,
0.0230255126953125,
-0.0467529296875,
0.03228759765625,
0.05291748046875,
-0.0140838623046875,
0.039825439453125,
-0.0008254051208496094,
-0.01165008544921875,
-0.0810546875,
0.002956390380859375,
-0.0157470703125,
0.0036449432373046875,
-0.03192138671875,
-0.00634765625,
-0.0157012939453125,
0.00690460205078125,
-0.046875,
0.04443359375,
-0.024993896484375,
-0.0113677978515625,
-0.009796142578125,
0.0020694732666015625,
0.0038280487060546875,
0.048431396484375,
-0.01210784912109375,
0.07965087890625,
0.029022216796875,
-0.0435791015625,
0.0182342529296875,
0.0290069580078125,
-0.037353515625,
0.00975799560546875,
-0.0662841796875,
0.026092529296875,
0.0081634521484375,
0.03936767578125,
-0.0748291015625,
-0.03021240234375,
0.0245513916015625,
-0.03448486328125,
0.007396697998046875,
0.0157928466796875,
-0.04180908203125,
-0.0308380126953125,
-0.031646728515625,
0.02392578125,
0.060516357421875,
-0.032928466796875,
0.011383056640625,
0.0277252197265625,
0.0014944076538085938,
-0.050689697265625,
-0.061920166015625,
0.00467681884765625,
-0.027374267578125,
-0.04058837890625,
0.0225372314453125,
-0.0135955810546875,
-0.0175933837890625,
-0.0194854736328125,
0.00606536865234375,
-0.00015473365783691406,
0.028717041015625,
0.0289764404296875,
0.0289764404296875,
-0.00926971435546875,
-0.0015325546264648438,
0.010894775390625,
-0.013336181640625,
0.003021240234375,
0.01366424560546875,
0.04705810546875,
-0.01386260986328125,
-0.0168914794921875,
-0.05706787109375,
0.0013427734375,
0.022216796875,
-0.018096923828125,
0.04705810546875,
0.031829833984375,
-0.017669677734375,
0.0199127197265625,
-0.059173583984375,
-0.0085296630859375,
-0.0401611328125,
0.040374755859375,
-0.01456451416015625,
-0.06243896484375,
0.040313720703125,
0.002338409423828125,
0.029815673828125,
0.056976318359375,
0.04803466796875,
-0.006103515625,
0.058441162109375,
0.0406494140625,
-0.0029010772705078125,
0.0259552001953125,
-0.03753662109375,
-0.00921630859375,
-0.070556640625,
-0.047149658203125,
-0.0221099853515625,
-0.030731201171875,
-0.049530029296875,
-0.0321044921875,
0.0199737548828125,
0.0153350830078125,
-0.052581787109375,
0.0228271484375,
-0.0423583984375,
0.041534423828125,
0.0404052734375,
0.0118408203125,
0.0221099853515625,
0.006305694580078125,
0.00937652587890625,
0.004772186279296875,
-0.040191650390625,
-0.055938720703125,
0.1119384765625,
0.033599853515625,
0.03424072265625,
0.0077972412109375,
0.051422119140625,
0.01209259033203125,
0.0237579345703125,
-0.0517578125,
0.046875,
0.004611968994140625,
-0.0545654296875,
-0.01483154296875,
-0.0080413818359375,
-0.06842041015625,
0.01094818115234375,
-0.0176849365234375,
-0.057159423828125,
0.0026149749755859375,
0.00009381771087646484,
-0.02996826171875,
0.0199737548828125,
-0.052978515625,
0.0458984375,
-0.042083740234375,
-0.0243377685546875,
-0.0243377685546875,
-0.061553955078125,
0.0506591796875,
-0.0132293701171875,
0.0090789794921875,
-0.037139892578125,
-0.0172119140625,
0.0704345703125,
-0.026611328125,
0.0770263671875,
-0.003330230712890625,
-0.007442474365234375,
0.044586181640625,
-0.01374053955078125,
0.033233642578125,
0.003726959228515625,
-0.01898193359375,
0.04937744140625,
-0.01043701171875,
-0.0246734619140625,
-0.011962890625,
0.039794921875,
-0.091552734375,
-0.060211181640625,
-0.035675048828125,
-0.03729248046875,
-0.0023746490478515625,
0.007488250732421875,
0.0382080078125,
-0.005886077880859375,
-0.0015201568603515625,
0.0108184814453125,
0.034759521484375,
-0.038116455078125,
0.0350341796875,
0.04144287109375,
-0.00556182861328125,
-0.037200927734375,
0.050323486328125,
0.004055023193359375,
0.028076171875,
0.0175628662109375,
0.003936767578125,
-0.02996826171875,
-0.032440185546875,
-0.037933349609375,
0.0220489501953125,
-0.03558349609375,
-0.036041259765625,
-0.040985107421875,
-0.0250396728515625,
-0.02435302734375,
-0.004119873046875,
-0.034027099609375,
-0.031829833984375,
-0.06005859375,
-0.0284881591796875,
0.037139892578125,
0.06134033203125,
0.0004451274871826172,
0.047515869140625,
-0.0259552001953125,
0.01300811767578125,
0.0275115966796875,
0.0147705078125,
0.0004436969757080078,
-0.058990478515625,
0.0023822784423828125,
0.01157379150390625,
-0.0574951171875,
-0.047332763671875,
0.018646240234375,
0.019866943359375,
0.035797119140625,
0.03607177734375,
-0.00457000732421875,
0.057830810546875,
-0.025299072265625,
0.08221435546875,
0.0283203125,
-0.048370361328125,
0.052215576171875,
-0.0168609619140625,
0.004741668701171875,
0.046142578125,
0.01837158203125,
-0.00457000732421875,
-0.01180267333984375,
-0.049530029296875,
-0.051422119140625,
0.05828857421875,
0.0175628662109375,
0.01259613037109375,
0.0036029815673828125,
0.034820556640625,
0.00624847412109375,
0.0077972412109375,
-0.062255859375,
-0.023345947265625,
-0.021636962890625,
-0.00921630859375,
-0.01552581787109375,
-0.037872314453125,
-0.0038089752197265625,
-0.024871826171875,
0.0472412109375,
0.0038051605224609375,
0.0282440185546875,
-0.00904083251953125,
0.001064300537109375,
-0.00677490234375,
0.006008148193359375,
0.0555419921875,
0.036956787109375,
-0.0201568603515625,
-0.01027679443359375,
0.048248291015625,
-0.048095703125,
0.0263824462890625,
0.0017671585083007812,
-0.00893402099609375,
-0.0299224853515625,
0.030670166015625,
0.06781005859375,
0.019378662109375,
-0.05377197265625,
0.025604248046875,
0.010345458984375,
-0.027679443359375,
-0.030548095703125,
0.0283355712890625,
0.00768280029296875,
0.0253448486328125,
0.0221099853515625,
-0.01190185546875,
0.00691986083984375,
-0.037933349609375,
-0.0095977783203125,
0.02783203125,
0.007610321044921875,
-0.030731201171875,
0.074951171875,
0.023773193359375,
-0.02099609375,
0.03961181640625,
-0.014862060546875,
-0.0271148681640625,
0.06768798828125,
0.04931640625,
0.0458984375,
-0.0167694091796875,
0.0088348388671875,
0.05328369140625,
0.033782958984375,
-0.01629638671875,
0.016571044921875,
-0.000028371810913085938,
-0.036773681640625,
-0.0145721435546875,
-0.052642822265625,
-0.0340576171875,
0.026641845703125,
-0.0416259765625,
0.02294921875,
-0.0465087890625,
-0.0208282470703125,
-0.0242767333984375,
0.035888671875,
-0.05194091796875,
0.017425537109375,
0.00867462158203125,
0.0677490234375,
-0.053924560546875,
0.057830810546875,
0.036773681640625,
-0.03857421875,
-0.06658935546875,
-0.02288818359375,
0.01334381103515625,
-0.0908203125,
0.03863525390625,
0.0282135009765625,
-0.003330230712890625,
0.00991058349609375,
-0.05743408203125,
-0.091796875,
0.1297607421875,
0.0364990234375,
-0.058013916015625,
-0.001941680908203125,
0.0233154296875,
0.0377197265625,
-0.007472991943359375,
0.032745361328125,
0.061553955078125,
0.035888671875,
0.009979248046875,
-0.079833984375,
0.00742340087890625,
-0.0269012451171875,
-0.0037326812744140625,
-0.0148162841796875,
-0.09814453125,
0.06195068359375,
-0.031494140625,
-0.0173187255859375,
0.01416015625,
0.050689697265625,
0.0513916015625,
0.0400390625,
0.025390625,
0.057952880859375,
0.06951904296875,
-0.00139617919921875,
0.0819091796875,
-0.0255584716796875,
0.0140228271484375,
0.06744384765625,
-0.0206146240234375,
0.07415771484375,
0.01885986328125,
-0.044708251953125,
0.0450439453125,
0.07666015625,
-0.0014390945434570312,
0.045074462890625,
0.004608154296875,
-0.01076507568359375,
-0.01335906982421875,
-0.01171875,
-0.047698974609375,
0.0374755859375,
0.0193939208984375,
-0.01024627685546875,
-0.0006685256958007812,
-0.0241241455078125,
0.0159149169921875,
-0.025299072265625,
-0.0005502700805664062,
0.060302734375,
0.01122283935546875,
-0.046417236328125,
0.0697021484375,
0.0038051605224609375,
0.0650634765625,
-0.046234130859375,
0.006687164306640625,
-0.038787841796875,
0.0006275177001953125,
-0.0273284912109375,
-0.053375244140625,
0.003971099853515625,
0.02606201171875,
-0.0012769699096679688,
-0.003662109375,
0.040924072265625,
0.0025539398193359375,
-0.04205322265625,
0.02545166015625,
0.0207061767578125,
0.0255889892578125,
0.01605224609375,
-0.048675537109375,
0.013580322265625,
0.00698089599609375,
-0.039581298828125,
0.0283050537109375,
0.0028705596923828125,
-0.00476837158203125,
0.060882568359375,
0.055450439453125,
-0.015869140625,
0.01161956787109375,
-0.0171356201171875,
0.07391357421875,
-0.03741455078125,
-0.0165863037109375,
-0.05706787109375,
0.039215087890625,
0.0050506591796875,
-0.05340576171875,
0.03814697265625,
0.0477294921875,
0.052642822265625,
0.020233154296875,
0.049957275390625,
0.004505157470703125,
0.0230560302734375,
-0.040802001953125,
0.045623779296875,
-0.05755615234375,
0.0301513671875,
0.005832672119140625,
-0.07135009765625,
-0.0051422119140625,
0.04949951171875,
-0.017059326171875,
0.0028324127197265625,
0.02801513671875,
0.0645751953125,
0.01251983642578125,
-0.01274871826171875,
0.0097503662109375,
0.01450347900390625,
0.0275115966796875,
0.06622314453125,
0.0614013671875,
-0.047454833984375,
0.053741455078125,
-0.0297393798828125,
-0.017791748046875,
-0.0223541259765625,
-0.053924560546875,
-0.072998046875,
-0.0209808349609375,
-0.0159149169921875,
-0.0118408203125,
0.005908966064453125,
0.0594482421875,
0.0357666015625,
-0.043182373046875,
-0.023193359375,
-0.004863739013671875,
-0.006687164306640625,
0.002788543701171875,
-0.01194000244140625,
0.0268096923828125,
-0.006488800048828125,
-0.04144287109375,
0.033782958984375,
-0.0007081031799316406,
0.01849365234375,
-0.0254669189453125,
-0.022308349609375,
-0.0171051025390625,
0.01041412353515625,
0.046539306640625,
0.020782470703125,
-0.0703125,
-0.0167694091796875,
0.003925323486328125,
-0.01261138916015625,
0.01049041748046875,
-0.0026493072509765625,
-0.0576171875,
0.00562286376953125,
0.0098114013671875,
0.0285186767578125,
0.048736572265625,
0.005428314208984375,
0.00405120849609375,
-0.036956787109375,
0.033660888671875,
0.0004391670227050781,
0.01104736328125,
0.0251617431640625,
-0.03240966796875,
0.05657958984375,
0.01218414306640625,
-0.051727294921875,
-0.0699462890625,
0.00601959228515625,
-0.08026123046875,
-0.0009250640869140625,
0.10552978515625,
0.00128173828125,
-0.01050567626953125,
0.013916015625,
-0.0189361572265625,
0.03021240234375,
-0.0302581787109375,
0.06048583984375,
0.042694091796875,
-0.006122589111328125,
-0.00664520263671875,
-0.057861328125,
0.0247650146484375,
0.0277252197265625,
-0.08306884765625,
-0.0182952880859375,
0.035858154296875,
0.03765869140625,
-0.007965087890625,
0.053009033203125,
0.00267791748046875,
0.01934814453125,
0.00540924072265625,
0.0076751708984375,
-0.020477294921875,
-0.01251220703125,
-0.005764007568359375,
-0.0178985595703125,
-0.000728607177734375,
-0.017242431640625
]
] |
heegyu/WizardVicuna2-13b-hf | 2023-08-07T12:16:04.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"facebook",
"meta",
"llama-2",
"en",
"dataset:ehartford/wizard_vicuna_70k_unfiltered",
"arxiv:2307.09288",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | heegyu | null | null | heegyu/WizardVicuna2-13b-hf | 0 | 7,543 | transformers | 2023-08-07T11:35:08 | ---
extra_gated_heading: Access Llama 2 on Hugging Face
extra_gated_description: >-
This is a form to enable access to Llama 2 on Hugging Face after you have been
granted access from Meta. Please visit the [Meta
website](https://ai.meta.com/resources/models-and-libraries/llama-downloads)
and accept our license terms and acceptable use policy before submitting this
form. Requests will be processed in 1-2 days.
extra_gated_prompt: >-
**Your Hugging Face account email address MUST match the email you provide on
the Meta website, or your request will not be approved.**
extra_gated_button_content: Submit
extra_gated_fields:
I agree to share my name, email address and username with Meta and confirm that I have already been granted download access on the Meta website: checkbox
language:
- en
pipeline_tag: text-generation
inference: false
tags:
- facebook
- meta
- pytorch
- llama
- llama-2
datasets:
- ehartford/wizard_vicuna_70k_unfiltered
---
finetuned Llama-2-13b-hf model with ehartford/wizard_vicuna_70k_unfiltered data (3 epoch)
### prompt
```
### Human:
Hi, how are you?
### Assistant:
```
# **Llama 2**
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 7B pretrained model. Links to other models can be found in the index at the bottom.
## Model Details
*Note: Use of this model is governed by the Meta license. 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.*
Meta developed and publicly released the Llama 2 family of large language models (LLMs), a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama-2-Chat, are optimized for dialogue use cases. Llama-2-Chat models outperform open-source chat models on most benchmarks we tested, and in our human evaluations for helpfulness and safety, are on par with some popular closed-source models like ChatGPT and PaLM.
**Model Developers** Meta
**Variations** Llama 2 comes in a range of parameter sizes — 7B, 13B, and 70B — as well as pretrained and fine-tuned variations.
**Input** Models input text only.
**Output** Models generate text only.
**Model Architecture** Llama 2 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.
||Training Data|Params|Content Length|GQA|Tokens|LR|
|---|---|---|---|---|---|---|
|Llama 2|*A new mix of publicly available online data*|7B|4k|✗|2.0T|3.0 x 10<sup>-4</sup>|
|Llama 2|*A new mix of publicly available online data*|13B|4k|✗|2.0T|3.0 x 10<sup>-4</sup>|
|Llama 2|*A new mix of publicly available online data*|70B|4k|✔|2.0T|1.5 x 10<sup>-4</sup>|
*Llama 2 family of models.* Token counts refer to pretraining data only. All models are trained with a global batch-size of 4M tokens. Bigger models - 70B -- use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Dates** Llama 2 was trained between January 2023 and July 2023.
**Status** This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.
**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** ["Llama-2: Open Foundation and Fine-tuned Chat Models"](arxiv.org/abs/2307.09288)
## Intended Use
**Intended Use Cases** Llama 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
To get the expected features and performance for the chat versions, a specific formatting needs to be followed, including the `INST` and `<<SYS>>` tags, `BOS` and `EOS` tokens, and the whitespaces and breaklines in between (we recommend calling `strip()` on inputs to avoid double-spaces). See our reference code in github for details: [`chat_completion`](https://github.com/facebookresearch/llama/blob/main/llama/generation.py#L212).
**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 Llama 2.
## Hardware and Software
**Training Factors** We used custom training libraries, Meta's Research Super Cluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.
**Carbon Footprint** Pretraining utilized a cumulative 3.3M GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 539 tCO2eq, 100% of which were offset by Meta’s sustainability program.
||Time (GPU hours)|Power Consumption (W)|Carbon Emitted(tCO<sub>2</sub>eq)|
|---|---|---|---|
|Llama 2 7B|184320|400|31.22|
|Llama 2 13B|368640|400|62.44|
|Llama 2 70B|1720320|400|291.42|
|Total|3311616||539.00|
**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** Llama 2 was pretrained on 2 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over one million new human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.
**Data Freshness** The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023.
## Evaluation Results
In this section, we report the results for the Llama 1 and Llama 2 models on standard academic benchmarks.For all the evaluations, we use our internal evaluations library.
|Model|Size|Code|Commonsense Reasoning|World Knowledge|Reading Comprehension|Math|MMLU|BBH|AGI Eval|
|---|---|---|---|---|---|---|---|---|---|
|Llama 1|7B|14.1|60.8|46.2|58.5|6.95|35.1|30.3|23.9|
|Llama 1|13B|18.9|66.1|52.6|62.3|10.9|46.9|37.0|33.9|
|Llama 1|33B|26.0|70.0|58.4|67.6|21.4|57.8|39.8|41.7|
|Llama 1|65B|30.7|70.7|60.5|68.6|30.8|63.4|43.5|47.6|
|Llama 2|7B|16.8|63.9|48.9|61.3|14.6|45.3|32.6|29.3|
|Llama 2|13B|24.5|66.9|55.4|65.8|28.7|54.8|39.4|39.1|
|Llama 2|70B|**37.5**|**71.9**|**63.6**|**69.4**|**35.2**|**68.9**|**51.2**|**54.2**|
**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
Llama 2 is 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’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, 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/](https://ai.meta.com/llama/responsible-use-guide)
## Reporting Issues
Please report any software “bug,” or other problems with the models through one of the following means:
- Reporting issues with the model: [github.com/facebookresearch/llama](http://github.com/facebookresearch/llama)
- Reporting problematic content generated by the model: [developers.facebook.com/llama_output_feedback](http://developers.facebook.com/llama_output_feedback)
- Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)
## Llama Model Index
|Model|Llama2|Llama2-hf|Llama2-chat|Llama2-chat-hf|
|---|---|---|---|---|
|7B| [Link](https://huggingface.co/llamaste/Llama-2-7b) | [Link](https://huggingface.co/llamaste/Llama-2-7b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat-hf)|
|13B| [Link](https://huggingface.co/llamaste/Llama-2-13b) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-13b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf)|
|70B| [Link](https://huggingface.co/llamaste/Llama-2-70b) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-70b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf)| | 10,520 | [
[
-0.0169525146484375,
-0.054931640625,
0.0283050537109375,
0.01256561279296875,
-0.02813720703125,
0.0160064697265625,
-0.00511932373046875,
-0.056732177734375,
0.004924774169921875,
0.0221099853515625,
-0.05401611328125,
-0.04248046875,
-0.048492431640625,
0.00827789306640625,
-0.0146331787109375,
0.0772705078125,
0.002620697021484375,
-0.020904541015625,
-0.0088348388671875,
0.00846099853515625,
-0.03631591796875,
-0.0309295654296875,
-0.039398193359375,
-0.034423828125,
0.0286102294921875,
0.038177490234375,
0.04534912109375,
0.050506591796875,
0.04052734375,
0.0180816650390625,
-0.019989013671875,
0.0178680419921875,
-0.055206298828125,
-0.0220947265625,
0.011016845703125,
-0.034698486328125,
-0.049774169921875,
0.01035308837890625,
0.02777099609375,
0.01568603515625,
-0.023101806640625,
0.0382080078125,
0.00705718994140625,
0.035186767578125,
-0.0416259765625,
0.0136260986328125,
-0.054718017578125,
0.0016994476318359375,
-0.0168609619140625,
-0.0068359375,
-0.0149078369140625,
-0.0194854736328125,
-0.01371002197265625,
-0.06378173828125,
-0.00888824462890625,
0.007694244384765625,
0.08013916015625,
0.048583984375,
-0.0321044921875,
-0.00893402099609375,
-0.02252197265625,
0.0684814453125,
-0.06640625,
0.006008148193359375,
0.04534912109375,
0.019989013671875,
-0.019561767578125,
-0.05633544921875,
-0.049560546875,
-0.01216888427734375,
0.005229949951171875,
0.0258636474609375,
-0.0301055908203125,
0.0015974044799804688,
0.01074981689453125,
0.028076171875,
-0.042877197265625,
0.043487548828125,
-0.038970947265625,
-0.01502227783203125,
0.0772705078125,
0.0193328857421875,
-0.0024967193603515625,
-0.00673675537109375,
-0.036468505859375,
-0.01971435546875,
-0.05963134765625,
0.01300048828125,
0.037750244140625,
-0.0012979507446289062,
-0.03485107421875,
0.0474853515625,
-0.0322265625,
0.02288818359375,
0.004688262939453125,
-0.0382080078125,
0.0367431640625,
-0.0352783203125,
-0.02252197265625,
-0.009033203125,
0.06817626953125,
0.053466796875,
0.0096893310546875,
0.00878143310546875,
-0.004100799560546875,
0.00662994384765625,
-0.0021762847900390625,
-0.06304931640625,
-0.005039215087890625,
0.0183258056640625,
-0.0287628173828125,
-0.04254150390625,
-0.02099609375,
-0.0548095703125,
-0.01336669921875,
-0.0078277587890625,
0.019195556640625,
-0.003025054931640625,
-0.02825927734375,
0.00974273681640625,
0.0052032470703125,
0.040679931640625,
0.0163726806640625,
-0.06976318359375,
0.0179901123046875,
0.03997802734375,
0.057525634765625,
-0.018402099609375,
-0.0266571044921875,
0.0029659271240234375,
-0.002666473388671875,
-0.0242462158203125,
0.06689453125,
-0.024688720703125,
-0.0382080078125,
-0.01806640625,
-0.0014257431030273438,
0.01331329345703125,
-0.0396728515625,
0.033538818359375,
-0.0303802490234375,
0.01473236083984375,
-0.02392578125,
-0.0290985107421875,
-0.025238037109375,
0.01513671875,
-0.030670166015625,
0.10894775390625,
0.00905609130859375,
-0.037811279296875,
0.0231475830078125,
-0.050262451171875,
-0.01318359375,
-0.01483154296875,
0.007965087890625,
-0.040191650390625,
-0.0237884521484375,
0.010101318359375,
0.029266357421875,
-0.049468994140625,
0.036590576171875,
-0.0175018310546875,
-0.033782958984375,
0.004337310791015625,
-0.035797119140625,
0.06396484375,
0.022796630859375,
-0.033416748046875,
0.003170013427734375,
-0.06488037109375,
0.0034198760986328125,
0.034454345703125,
-0.038055419921875,
0.0196075439453125,
0.0030517578125,
-0.0076904296875,
0.015533447265625,
0.038299560546875,
-0.028106689453125,
0.00986480712890625,
-0.0258941650390625,
0.037017822265625,
0.05560302734375,
0.0049591064453125,
0.0128173828125,
-0.03851318359375,
0.03759765625,
-0.002979278564453125,
0.0298614501953125,
0.00180816650390625,
-0.054473876953125,
-0.0758056640625,
-0.0146026611328125,
-0.00162506103515625,
0.062103271484375,
-0.018768310546875,
0.05389404296875,
-0.00021147727966308594,
-0.055816650390625,
-0.033111572265625,
0.028228759765625,
0.050048828125,
0.038665771484375,
0.0323486328125,
-0.0206756591796875,
-0.0458984375,
-0.07666015625,
0.007389068603515625,
-0.033966064453125,
-0.0018110275268554688,
0.0277099609375,
0.04803466796875,
-0.02520751953125,
0.0552978515625,
-0.0428466796875,
-0.01380157470703125,
-0.019622802734375,
-0.01038360595703125,
0.0059356689453125,
0.0261383056640625,
0.046539306640625,
-0.028472900390625,
-0.0164947509765625,
-0.00916290283203125,
-0.06658935546875,
-0.007755279541015625,
0.00824737548828125,
-0.0163421630859375,
0.01910400390625,
0.0236053466796875,
-0.046539306640625,
0.0325927734375,
0.052581787109375,
-0.01470184326171875,
0.039398193359375,
-0.0012111663818359375,
-0.011199951171875,
-0.08160400390625,
0.002910614013671875,
-0.01507568359375,
0.003265380859375,
-0.033477783203125,
-0.005573272705078125,
-0.01557159423828125,
0.00643157958984375,
-0.04669189453125,
0.04437255859375,
-0.024444580078125,
-0.00984954833984375,
-0.010406494140625,
0.0016832351684570312,
0.0035400390625,
0.04925537109375,
-0.01137542724609375,
0.07916259765625,
0.0294342041015625,
-0.0439453125,
0.0180511474609375,
0.02947998046875,
-0.03704833984375,
0.01004791259765625,
-0.0650634765625,
0.0259857177734375,
0.0084075927734375,
0.03900146484375,
-0.074462890625,
-0.0304718017578125,
0.024627685546875,
-0.03436279296875,
0.007610321044921875,
0.016632080078125,
-0.04156494140625,
-0.03106689453125,
-0.03173828125,
0.0224456787109375,
0.059967041015625,
-0.034088134765625,
0.0129241943359375,
0.02716064453125,
0.0025730133056640625,
-0.050689697265625,
-0.0625,
0.004558563232421875,
-0.0268402099609375,
-0.0411376953125,
0.023895263671875,
-0.0134735107421875,
-0.0179443359375,
-0.018463134765625,
0.005908966064453125,
-0.0015087127685546875,
0.028717041015625,
0.0282745361328125,
0.0291748046875,
-0.00847625732421875,
-0.0011186599731445312,
0.01013946533203125,
-0.01395416259765625,
0.00286865234375,
0.01468658447265625,
0.046295166015625,
-0.0136260986328125,
-0.0171051025390625,
-0.058197021484375,
0.0011510848999023438,
0.02142333984375,
-0.0172119140625,
0.04840087890625,
0.031768798828125,
-0.018463134765625,
0.0189666748046875,
-0.05828857421875,
-0.009735107421875,
-0.0404052734375,
0.041046142578125,
-0.01445770263671875,
-0.0625,
0.04022216796875,
0.0022563934326171875,
0.0298614501953125,
0.0577392578125,
0.0474853515625,
-0.006626129150390625,
0.058502197265625,
0.04022216796875,
-0.00304412841796875,
0.025634765625,
-0.03741455078125,
-0.0086517333984375,
-0.069091796875,
-0.0462646484375,
-0.02294921875,
-0.0299835205078125,
-0.050262451171875,
-0.030670166015625,
0.0200958251953125,
0.01470947265625,
-0.052459716796875,
0.022735595703125,
-0.041595458984375,
0.042083740234375,
0.040069580078125,
0.0123443603515625,
0.0233154296875,
0.007015228271484375,
0.01013946533203125,
0.004669189453125,
-0.040985107421875,
-0.0552978515625,
0.1124267578125,
0.03302001953125,
0.0345458984375,
0.006572723388671875,
0.051605224609375,
0.01213836669921875,
0.0234527587890625,
-0.052703857421875,
0.046630859375,
0.005214691162109375,
-0.055816650390625,
-0.016082763671875,
-0.00847625732421875,
-0.067626953125,
0.0114593505859375,
-0.0178680419921875,
-0.055694580078125,
0.0028362274169921875,
-0.0002008676528930664,
-0.030731201171875,
0.02020263671875,
-0.05224609375,
0.0460205078125,
-0.0406494140625,
-0.02410888671875,
-0.0254974365234375,
-0.059844970703125,
0.0506591796875,
-0.0149078369140625,
0.0087127685546875,
-0.036224365234375,
-0.017822265625,
0.07000732421875,
-0.0273895263671875,
0.076416015625,
-0.00299072265625,
-0.00827789306640625,
0.044464111328125,
-0.0145721435546875,
0.034027099609375,
0.0027103424072265625,
-0.01983642578125,
0.04888916015625,
-0.01001739501953125,
-0.023956298828125,
-0.0112457275390625,
0.040191650390625,
-0.09149169921875,
-0.059844970703125,
-0.0367431640625,
-0.036468505859375,
-0.0022602081298828125,
0.007236480712890625,
0.038299560546875,
-0.004924774169921875,
-0.0022525787353515625,
0.0106201171875,
0.034088134765625,
-0.038360595703125,
0.03533935546875,
0.0426025390625,
-0.0056304931640625,
-0.036651611328125,
0.0504150390625,
0.00437164306640625,
0.02783203125,
0.0174407958984375,
0.0043487548828125,
-0.0299224853515625,
-0.031890869140625,
-0.03741455078125,
0.021514892578125,
-0.036163330078125,
-0.035919189453125,
-0.04095458984375,
-0.0263671875,
-0.0228424072265625,
-0.0040283203125,
-0.033966064453125,
-0.032501220703125,
-0.0596923828125,
-0.02862548828125,
0.038055419921875,
0.06134033203125,
-0.00022530555725097656,
0.04705810546875,
-0.0253448486328125,
0.0134735107421875,
0.0283355712890625,
0.0146026611328125,
0.0010557174682617188,
-0.06036376953125,
0.0013265609741210938,
0.01085662841796875,
-0.057952880859375,
-0.047149658203125,
0.0189056396484375,
0.0216217041015625,
0.03631591796875,
0.036102294921875,
-0.0038204193115234375,
0.058349609375,
-0.02606201171875,
0.08197021484375,
0.0271148681640625,
-0.047637939453125,
0.050567626953125,
-0.0173187255859375,
0.005016326904296875,
0.046783447265625,
0.0190887451171875,
-0.0048828125,
-0.0121002197265625,
-0.0484619140625,
-0.050811767578125,
0.059844970703125,
0.0178680419921875,
0.01241302490234375,
0.0031375885009765625,
0.034942626953125,
0.005825042724609375,
0.006412506103515625,
-0.0625,
-0.0245361328125,
-0.0205841064453125,
-0.009002685546875,
-0.01433563232421875,
-0.03680419921875,
-0.0038204193115234375,
-0.0259246826171875,
0.048614501953125,
0.00411224365234375,
0.028076171875,
-0.00952911376953125,
0.00118255615234375,
-0.00836181640625,
0.005657196044921875,
0.05389404296875,
0.037811279296875,
-0.0204315185546875,
-0.0116729736328125,
0.04840087890625,
-0.048187255859375,
0.026031494140625,
0.0020618438720703125,
-0.00951385498046875,
-0.0304412841796875,
0.0306243896484375,
0.0670166015625,
0.0197906494140625,
-0.053253173828125,
0.0254974365234375,
0.0102081298828125,
-0.0287017822265625,
-0.031524658203125,
0.027130126953125,
0.0078125,
0.0246124267578125,
0.0226898193359375,
-0.01110076904296875,
0.007160186767578125,
-0.0377197265625,
-0.00899505615234375,
0.0282440185546875,
0.00849151611328125,
-0.030853271484375,
0.0740966796875,
0.0245208740234375,
-0.0209197998046875,
0.039825439453125,
-0.0143585205078125,
-0.0268707275390625,
0.0673828125,
0.048431396484375,
0.047210693359375,
-0.0179443359375,
0.009246826171875,
0.053802490234375,
0.033203125,
-0.0159912109375,
0.016815185546875,
0.0005345344543457031,
-0.038421630859375,
-0.01531219482421875,
-0.0518798828125,
-0.034088134765625,
0.0274658203125,
-0.04205322265625,
0.0227813720703125,
-0.045440673828125,
-0.020416259765625,
-0.023681640625,
0.035247802734375,
-0.0523681640625,
0.01666259765625,
0.00867462158203125,
0.06756591796875,
-0.054656982421875,
0.05792236328125,
0.036712646484375,
-0.037933349609375,
-0.0665283203125,
-0.021209716796875,
0.013824462890625,
-0.09075927734375,
0.03851318359375,
0.027862548828125,
-0.0038051605224609375,
0.01044464111328125,
-0.056915283203125,
-0.09228515625,
0.1297607421875,
0.034942626953125,
-0.057220458984375,
-0.00196075439453125,
0.023406982421875,
0.03814697265625,
-0.006977081298828125,
0.03302001953125,
0.062225341796875,
0.03680419921875,
0.00975799560546875,
-0.0797119140625,
0.00662994384765625,
-0.0262298583984375,
-0.00443267822265625,
-0.01531219482421875,
-0.0977783203125,
0.06134033203125,
-0.0295257568359375,
-0.0183258056640625,
0.0148773193359375,
0.050506591796875,
0.051300048828125,
0.039947509765625,
0.0255126953125,
0.058868408203125,
0.06976318359375,
-0.001003265380859375,
0.082275390625,
-0.02655029296875,
0.013275146484375,
0.06719970703125,
-0.0208892822265625,
0.07427978515625,
0.0183258056640625,
-0.045379638671875,
0.0445556640625,
0.0775146484375,
-0.0006570816040039062,
0.046173095703125,
0.004276275634765625,
-0.010833740234375,
-0.01445770263671875,
-0.011474609375,
-0.047271728515625,
0.0377197265625,
0.018829345703125,
-0.01029205322265625,
-0.0014743804931640625,
-0.0233306884765625,
0.015167236328125,
-0.0243377685546875,
-0.001232147216796875,
0.060882568359375,
0.01134490966796875,
-0.046630859375,
0.07073974609375,
0.00447845458984375,
0.0660400390625,
-0.04669189453125,
0.006916046142578125,
-0.03857421875,
0.0009584426879882812,
-0.027435302734375,
-0.05303955078125,
0.004138946533203125,
0.026123046875,
-0.0013036727905273438,
-0.004619598388671875,
0.04156494140625,
0.0021266937255859375,
-0.04241943359375,
0.025787353515625,
0.021636962890625,
0.025390625,
0.0156707763671875,
-0.04937744140625,
0.01165008544921875,
0.007354736328125,
-0.040771484375,
0.028106689453125,
0.0031681060791015625,
-0.0037670135498046875,
0.060943603515625,
0.05438232421875,
-0.0163726806640625,
0.01155853271484375,
-0.016357421875,
0.0740966796875,
-0.037628173828125,
-0.0160064697265625,
-0.056915283203125,
0.0384521484375,
0.005634307861328125,
-0.0537109375,
0.0396728515625,
0.047271728515625,
0.0528564453125,
0.02001953125,
0.0489501953125,
0.0050811767578125,
0.02374267578125,
-0.04132080078125,
0.046295166015625,
-0.057830810546875,
0.0297088623046875,
0.005695343017578125,
-0.0726318359375,
-0.0044097900390625,
0.048828125,
-0.016510009765625,
0.003368377685546875,
0.0295257568359375,
0.0660400390625,
0.01291656494140625,
-0.01104736328125,
0.01036834716796875,
0.01451873779296875,
0.0262908935546875,
0.06536865234375,
0.0625,
-0.047210693359375,
0.05352783203125,
-0.0301055908203125,
-0.0180206298828125,
-0.021087646484375,
-0.054046630859375,
-0.0736083984375,
-0.021484375,
-0.01654052734375,
-0.0111541748046875,
0.005298614501953125,
0.060760498046875,
0.0361328125,
-0.044189453125,
-0.0228424072265625,
-0.0048828125,
-0.00823974609375,
0.00213623046875,
-0.01190948486328125,
0.027313232421875,
-0.006290435791015625,
-0.041961669921875,
0.035247802734375,
-0.0018062591552734375,
0.0175933837890625,
-0.025909423828125,
-0.02294921875,
-0.0177764892578125,
0.011322021484375,
0.046417236328125,
0.0208587646484375,
-0.0704345703125,
-0.016143798828125,
0.004009246826171875,
-0.01220703125,
0.00965118408203125,
-0.0029144287109375,
-0.057891845703125,
0.00539398193359375,
0.0102386474609375,
0.029022216796875,
0.049102783203125,
0.0054779052734375,
0.004047393798828125,
-0.036834716796875,
0.032684326171875,
-0.000015974044799804688,
0.010711669921875,
0.02471923828125,
-0.032989501953125,
0.056640625,
0.011749267578125,
-0.0523681640625,
-0.06982421875,
0.005397796630859375,
-0.078369140625,
-0.0019350051879882812,
0.10546875,
0.0021533966064453125,
-0.01053619384765625,
0.01314544677734375,
-0.017730712890625,
0.029632568359375,
-0.030914306640625,
0.059722900390625,
0.0445556640625,
-0.006175994873046875,
-0.006805419921875,
-0.057891845703125,
0.025848388671875,
0.0282440185546875,
-0.08306884765625,
-0.0178070068359375,
0.0364990234375,
0.036651611328125,
-0.0091094970703125,
0.05218505859375,
0.002742767333984375,
0.0189208984375,
0.00443267822265625,
0.00771331787109375,
-0.0211334228515625,
-0.0127105712890625,
-0.0071563720703125,
-0.0185089111328125,
-0.0015277862548828125,
-0.0165557861328125
]
] |
heegyu/WizardVicuna-3B-0719 | 2023-08-09T12:08:44.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"en",
"dataset:heegyu/wizard_vicuna_70k_v2",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | heegyu | null | null | heegyu/WizardVicuna-3B-0719 | 0 | 7,542 | transformers | 2023-07-23T02:51:40 | ---
license: apache-2.0
language:
- en
datasets:
- heegyu/wizard_vicuna_70k_v2
---
Base Model: [openlm-research/open_llama_3b](https://huggingface.co/openlm-research/open_llama_3b)
Usage
```
### Human:
your instruction
### ASSISANT:
output will be generated and ended with <|endoftext|>
``` | 294 | [
[
-0.002445220947265625,
-0.06024169921875,
0.01513671875,
-0.0005326271057128906,
-0.046630859375,
0.000042378902435302734,
0.0230865478515625,
-0.0090179443359375,
0.0225830078125,
0.05474853515625,
-0.0557861328125,
-0.058074951171875,
-0.036834716796875,
0.0021514892578125,
-0.03375244140625,
0.08026123046875,
-0.004932403564453125,
0.00232696533203125,
-0.019256591796875,
-0.0029811859130859375,
-0.0213470458984375,
-0.062469482421875,
-0.0341796875,
-0.0164642333984375,
0.042572021484375,
0.032562255859375,
0.047821044921875,
0.043487548828125,
0.0318603515625,
0.00820159912109375,
-0.014434814453125,
0.01068878173828125,
-0.0528564453125,
-0.00177001953125,
0.004192352294921875,
-0.036407470703125,
-0.07366943359375,
-0.0037097930908203125,
0.0362548828125,
0.03192138671875,
-0.01171112060546875,
0.04351806640625,
-0.0238494873046875,
0.0267181396484375,
-0.04071044921875,
-0.00209808349609375,
-0.0169219970703125,
0.005062103271484375,
-0.03680419921875,
0.005977630615234375,
-0.0200958251953125,
-0.033660888671875,
-0.0230560302734375,
-0.0289154052734375,
-0.005123138427734375,
0.0272369384765625,
0.06402587890625,
0.01509857177734375,
-0.033935546875,
-0.0061187744140625,
-0.0328369140625,
0.055511474609375,
-0.066650390625,
0.021575927734375,
0.051727294921875,
0.0225830078125,
-0.0323486328125,
-0.06939697265625,
-0.0294189453125,
-0.021881103515625,
0.00870513916015625,
-0.007259368896484375,
-0.025970458984375,
-0.0172576904296875,
0.01139068603515625,
0.016571044921875,
-0.03582763671875,
-0.0027675628662109375,
-0.0596923828125,
-0.0059051513671875,
0.0430908203125,
0.0218048095703125,
0.02020263671875,
0.01499176025390625,
-0.0202178955078125,
0.0048828125,
-0.06610107421875,
-0.0176544189453125,
0.033966064453125,
0.0189208984375,
-0.03826904296875,
0.07647705078125,
-0.0180206298828125,
0.0548095703125,
-0.0194244384765625,
-0.00873565673828125,
0.042694091796875,
0.0055084228515625,
-0.0341796875,
-0.00800323486328125,
0.042449951171875,
0.040313720703125,
0.00853729248046875,
0.0038661956787109375,
-0.01299285888671875,
0.003757476806640625,
0.00603485107421875,
-0.056549072265625,
-0.0000941157341003418,
-0.0024166107177734375,
-0.058502197265625,
-0.0257415771484375,
0.004589080810546875,
-0.04205322265625,
0.00685882568359375,
-0.0036029815673828125,
0.0128173828125,
-0.0054931640625,
-0.025604248046875,
0.01117706298828125,
0.0173797607421875,
0.02459716796875,
0.0059814453125,
-0.047332763671875,
0.053741455078125,
0.0284881591796875,
0.05535888671875,
0.00052642822265625,
-0.019805908203125,
-0.033660888671875,
0.027252197265625,
-0.020599365234375,
0.02587890625,
-0.0190582275390625,
-0.053436279296875,
-0.0177154541015625,
0.0039520263671875,
0.0211029052734375,
-0.0236968994140625,
0.042083740234375,
-0.04071044921875,
0.041046142578125,
-0.01297760009765625,
-0.02252197265625,
-0.054595947265625,
0.0028133392333984375,
-0.05731201171875,
0.07232666015625,
0.032257080078125,
-0.06219482421875,
-0.0092926025390625,
-0.08758544921875,
-0.0125732421875,
-0.006771087646484375,
0.00765228271484375,
-0.024749755859375,
-0.0035915374755859375,
-0.025543212890625,
0.02227783203125,
-0.037139892578125,
0.002498626708984375,
-0.023101806640625,
-0.016845703125,
0.007061004638671875,
0.004756927490234375,
0.072998046875,
-0.0006260871887207031,
0.005680084228515625,
0.031829833984375,
-0.06109619140625,
-0.02740478515625,
0.04937744140625,
-0.017242431640625,
-0.01861572265625,
-0.0072174072265625,
0.00672149658203125,
-0.009796142578125,
0.04315185546875,
-0.054046630859375,
0.038238525390625,
-0.00437164306640625,
0.01751708984375,
0.05267333984375,
0.005008697509765625,
0.020599365234375,
-0.0197601318359375,
0.033935546875,
-0.01216888427734375,
0.0234527587890625,
0.02435302734375,
-0.055694580078125,
-0.050933837890625,
-0.022857666015625,
-0.0033512115478515625,
0.0362548828125,
0.0024433135986328125,
0.0272216796875,
0.026031494140625,
-0.052978515625,
-0.0207061767578125,
-0.0012187957763671875,
0.0311431884765625,
0.046783447265625,
0.0289459228515625,
-0.0276336669921875,
-0.08453369140625,
-0.06219482421875,
0.0161285400390625,
-0.026519775390625,
0.0002980232238769531,
-0.000579833984375,
0.06219482421875,
-0.05621337890625,
0.0550537109375,
-0.06298828125,
-0.006717681884765625,
0.0013723373413085938,
0.00814056396484375,
0.0252838134765625,
0.0460205078125,
0.040283203125,
-0.03961181640625,
-0.0091705322265625,
-0.0133209228515625,
-0.057373046875,
-0.0082244873046875,
0.0316162109375,
-0.073486328125,
0.021240234375,
0.0012521743774414062,
-0.06817626953125,
0.0548095703125,
0.0189361572265625,
-0.0477294921875,
0.05303955078125,
0.0267333984375,
-0.023223876953125,
-0.0718994140625,
0.01335906982421875,
-0.036956787109375,
-0.01837158203125,
-0.037506103515625,
0.006927490234375,
-0.025115966796875,
0.024871826171875,
-0.03289794921875,
0.044952392578125,
-0.03155517578125,
-0.03033447265625,
-0.0030994415283203125,
-0.0297393798828125,
0.006137847900390625,
0.028564453125,
-0.005741119384765625,
0.04119873046875,
0.0247955322265625,
-0.0235443115234375,
0.04046630859375,
0.05010986328125,
0.006771087646484375,
-0.00878143310546875,
-0.052154541015625,
0.0408935546875,
0.005016326904296875,
0.02642822265625,
-0.05047607421875,
-0.041259765625,
0.056884765625,
-0.0372314453125,
-0.005474090576171875,
0.01070404052734375,
-0.053497314453125,
-0.0290069580078125,
-0.0261688232421875,
0.036041259765625,
0.0673828125,
-0.01812744140625,
0.043853759765625,
0.01910400390625,
0.0020656585693359375,
-0.02777099609375,
-0.055206298828125,
-0.01282501220703125,
-0.005245208740234375,
-0.0277557373046875,
0.001201629638671875,
-0.0142669677734375,
-0.0249786376953125,
0.0000012516975402832031,
0.020111083984375,
-0.012359619140625,
-0.0021953582763671875,
0.0289154052734375,
0.053131103515625,
-0.037445068359375,
-0.0196990966796875,
-0.006855010986328125,
0.0120697021484375,
0.02197265625,
0.027252197265625,
0.048370361328125,
-0.01415252685546875,
-0.05908203125,
-0.0084991455078125,
0.037994384765625,
0.033172607421875,
-0.010772705078125,
0.039093017578125,
0.04071044921875,
-0.059478759765625,
0.02301025390625,
-0.0182037353515625,
0.0116424560546875,
-0.023468017578125,
0.0258941650390625,
-0.0022640228271484375,
-0.033477783203125,
0.034271240234375,
0.0187835693359375,
0.01812744140625,
0.037872314453125,
0.06341552734375,
0.0196685791015625,
0.078369140625,
0.04412841796875,
0.0211639404296875,
0.004253387451171875,
-0.0152587890625,
-0.0125732421875,
-0.09698486328125,
-0.0447998046875,
-0.0222625732421875,
0.00279998779296875,
0.0026493072509765625,
-0.04376220703125,
0.007419586181640625,
0.03436279296875,
-0.056121826171875,
0.030120849609375,
-0.037750244140625,
0.01263427734375,
0.056884765625,
0.026153564453125,
0.0174713134765625,
-0.016082763671875,
-0.00264739990234375,
0.008575439453125,
-0.0234222412109375,
-0.0662841796875,
0.0814208984375,
0.03765869140625,
0.049560546875,
0.0302276611328125,
0.06341552734375,
0.0151519775390625,
0.049957275390625,
-0.0298004150390625,
0.043487548828125,
0.004032135009765625,
-0.044219970703125,
-0.00594329833984375,
0.006572723388671875,
-0.061004638671875,
0.01139068603515625,
0.0106048583984375,
-0.057037353515625,
-0.01256561279296875,
0.01078033447265625,
-0.0059051513671875,
0.0219573974609375,
-0.028839111328125,
0.0543212890625,
-0.0306396484375,
-0.0086212158203125,
0.01059722900390625,
-0.04705810546875,
0.06414794921875,
0.006473541259765625,
0.007312774658203125,
-0.01824951171875,
-0.041107177734375,
0.06524658203125,
-0.02545166015625,
0.0836181640625,
-0.006866455078125,
0.0006723403930664062,
0.0164031982421875,
0.01282501220703125,
0.020660400390625,
-0.0204925537109375,
-0.00765228271484375,
0.0181732177734375,
-0.0343017578125,
-0.023162841796875,
0.007488250732421875,
0.020904541015625,
-0.061614990234375,
-0.0233917236328125,
-0.0224151611328125,
-0.028564453125,
0.01026153564453125,
0.01064300537109375,
0.0294189453125,
0.0258636474609375,
-0.0379638671875,
-0.0017910003662109375,
0.0173492431640625,
-0.0131378173828125,
0.01291656494140625,
0.0292205810546875,
-0.056793212890625,
-0.041839599609375,
0.037017822265625,
-0.0081634521484375,
0.0037441253662109375,
0.0208892822265625,
0.0178680419921875,
-0.03472900390625,
-0.0303955078125,
-0.044769287109375,
0.0248565673828125,
-0.06561279296875,
-0.0504150390625,
-0.04248046875,
-0.021209716796875,
-0.02093505859375,
-0.00939178466796875,
-0.0020084381103515625,
-0.0160369873046875,
-0.07452392578125,
-0.029388427734375,
0.05120849609375,
0.06793212890625,
-0.02325439453125,
0.045867919921875,
-0.0582275390625,
0.04046630859375,
0.01291656494140625,
0.01413726806640625,
0.0167388916015625,
-0.0276947021484375,
-0.0166778564453125,
0.006439208984375,
-0.038787841796875,
-0.0765380859375,
0.046905517578125,
0.01151275634765625,
0.056304931640625,
0.00861358642578125,
0.00445556640625,
0.03582763671875,
-0.0271148681640625,
0.06719970703125,
0.0286865234375,
-0.05816650390625,
0.04315185546875,
-0.0262298583984375,
0.010498046875,
0.00806427001953125,
0.0245819091796875,
-0.00666046142578125,
0.0003731250762939453,
-0.047088623046875,
-0.081787109375,
0.0265350341796875,
0.0302734375,
0.0029582977294921875,
-0.0136260986328125,
-0.006618499755859375,
0.033050537109375,
0.0292510986328125,
-0.0638427734375,
-0.025360107421875,
-0.0162353515625,
-0.007152557373046875,
0.0345458984375,
-0.0377197265625,
-0.025909423828125,
-0.0205230712890625,
0.059173583984375,
0.01291656494140625,
0.0216217041015625,
-0.004497528076171875,
-0.0016603469848632812,
-0.036346435546875,
-0.017852783203125,
0.057830810546875,
0.034393310546875,
-0.0440673828125,
-0.01535797119140625,
0.0206451416015625,
-0.02789306640625,
-0.0011844635009765625,
-0.0157012939453125,
0.0069427490234375,
0.0031147003173828125,
0.052459716796875,
0.06353759765625,
0.0030994415283203125,
-0.0391845703125,
0.0228729248046875,
0.00801849365234375,
-0.00618743896484375,
-0.057830810546875,
0.0007529258728027344,
0.01806640625,
0.0020465850830078125,
0.0294189453125,
0.0115509033203125,
0.0197906494140625,
-0.039459228515625,
0.007450103759765625,
0.01392364501953125,
0.0087127685546875,
-0.02081298828125,
0.05426025390625,
0.0186309814453125,
-0.043182373046875,
0.04083251953125,
-0.00977325439453125,
-0.0523681640625,
0.05767822265625,
0.0572509765625,
0.06585693359375,
-0.01519775390625,
-0.0264892578125,
0.0243682861328125,
0.0086822509765625,
0.00955963134765625,
0.0616455078125,
0.01479339599609375,
-0.032470703125,
-0.0271148681640625,
-0.056121826171875,
-0.043212890625,
0.007068634033203125,
-0.058258056640625,
0.03338623046875,
-0.03515625,
0.001560211181640625,
-0.022003173828125,
-0.0176544189453125,
-0.036407470703125,
0.00865936279296875,
0.0255126953125,
0.0950927734375,
-0.052154541015625,
0.04864501953125,
0.0731201171875,
-0.05389404296875,
-0.0447998046875,
-0.0255584716796875,
0.005157470703125,
-0.068359375,
0.06884765625,
0.00986480712890625,
-0.022430419921875,
-0.013946533203125,
-0.0909423828125,
-0.0792236328125,
0.0938720703125,
0.01806640625,
-0.032257080078125,
-0.00240325927734375,
-0.0266571044921875,
0.0236968994140625,
-0.059783935546875,
0.0338134765625,
0.0197296142578125,
0.031494140625,
0.0159149169921875,
-0.04742431640625,
0.0212860107421875,
-0.027618408203125,
0.00908660888671875,
0.0018568038940429688,
-0.044219970703125,
0.091796875,
-0.0240478515625,
-0.005298614501953125,
0.035888671875,
0.06103515625,
0.040924072265625,
0.016204833984375,
0.030609130859375,
0.034027099609375,
0.030120849609375,
0.0215911865234375,
0.054962158203125,
0.0183563232421875,
0.035125732421875,
0.08941650390625,
-0.043792724609375,
0.0738525390625,
0.0265655517578125,
-0.0283050537109375,
0.07550048828125,
0.059906005859375,
-0.0200958251953125,
0.032501220703125,
0.03265380859375,
-0.007076263427734375,
0.00798797607421875,
-0.00959014892578125,
-0.051483154296875,
0.04248046875,
0.032867431640625,
-0.027313232421875,
-0.0072479248046875,
-0.0209503173828125,
0.0264892578125,
-0.00543212890625,
-0.02398681640625,
0.04815673828125,
0.003704071044921875,
-0.020294189453125,
0.056304931640625,
0.0048980712890625,
0.053619384765625,
-0.04864501953125,
-0.042083740234375,
-0.0020599365234375,
-0.005786895751953125,
-0.0309906005859375,
-0.0465087890625,
0.040252685546875,
0.01140594482421875,
-0.0299530029296875,
0.0147552490234375,
0.06109619140625,
-0.0167694091796875,
-0.039520263671875,
0.03802490234375,
0.03826904296875,
0.0435791015625,
0.01824951171875,
-0.044189453125,
0.026123046875,
0.017730712890625,
-0.044189453125,
0.0173797607421875,
0.00856781005859375,
-0.01131439208984375,
0.049957275390625,
0.07879638671875,
-0.00550079345703125,
0.0019025802612304688,
0.003284454345703125,
0.085205078125,
-0.046600341796875,
-0.018402099609375,
-0.07025146484375,
0.04241943359375,
-0.002582550048828125,
-0.042938232421875,
0.0269622802734375,
0.055328369140625,
0.0513916015625,
-0.0263824462890625,
0.05615234375,
-0.02886962890625,
0.0545654296875,
-0.028228759765625,
0.040863037109375,
-0.032501220703125,
-0.01212310791015625,
0.0077362060546875,
-0.05889892578125,
-0.0200653076171875,
0.05499267578125,
0.0185546875,
-0.0011224746704101562,
0.027862548828125,
0.054046630859375,
-0.01544952392578125,
-0.00609588623046875,
0.0257110595703125,
0.01641845703125,
-0.01119232177734375,
0.0233154296875,
0.049713134765625,
-0.0213470458984375,
0.010589599609375,
-0.029632568359375,
-0.03631591796875,
-0.0286102294921875,
-0.08013916015625,
-0.062286376953125,
-0.020111083984375,
-0.00445556640625,
-0.035980224609375,
-0.0005612373352050781,
0.088623046875,
0.052337646484375,
-0.056549072265625,
-0.045867919921875,
0.00429534912109375,
0.0127410888671875,
0.0012722015380859375,
-0.01094818115234375,
0.023529052734375,
0.03338623046875,
-0.03369140625,
-0.004657745361328125,
-0.01264190673828125,
0.0303955078125,
-0.03656005859375,
-0.0302276611328125,
-0.039093017578125,
-0.00977325439453125,
0.052581787109375,
0.020965576171875,
-0.0797119140625,
-0.0012655258178710938,
-0.0042266845703125,
-0.006191253662109375,
-0.009857177734375,
0.01326751708984375,
-0.038330078125,
-0.002971649169921875,
0.031951904296875,
0.009857177734375,
0.02191162109375,
0.009307861328125,
0.060882568359375,
-0.05267333984375,
-0.00003921985626220703,
0.0236053466796875,
0.045745849609375,
0.04144287109375,
-0.0433349609375,
0.0731201171875,
0.007068634033203125,
-0.034454345703125,
-0.07257080078125,
0.01534271240234375,
-0.07550048828125,
0.0013637542724609375,
0.08551025390625,
-0.01464080810546875,
-0.0491943359375,
0.016845703125,
-0.058319091796875,
0.055084228515625,
-0.048248291015625,
0.0179595947265625,
0.029327392578125,
-0.0299835205078125,
-0.015472412109375,
-0.037872314453125,
0.00024080276489257812,
-0.007625579833984375,
-0.0626220703125,
-0.01232147216796875,
0.036041259765625,
0.013458251953125,
0.0200653076171875,
0.051788330078125,
-0.00350189208984375,
0.0180206298828125,
0.0101318359375,
0.04632568359375,
-0.0259857177734375,
0.00798797607421875,
-0.0063018798828125,
-0.01219940185546875,
0.0097808837890625,
-0.055389404296875
]
] |
teknium/OpenHermes-7B | 2023-09-24T11:03:27.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-2",
"instruct",
"finetune",
"alpaca",
"gpt4",
"synthetic data",
"distillation",
"en",
"dataset:teknium/openhermes",
"license:mit",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | teknium | null | null | teknium/OpenHermes-7B | 9 | 7,540 | transformers | 2023-09-14T10:22:23 | ---
base_model: NousResearch/Llama-2-7b-hf
tags:
- llama-2
- instruct
- finetune
- alpaca
- gpt4
- synthetic data
- distillation
datasets:
- teknium/openhermes
model-index:
- name: openhermes-7b
results: []
license: mit
language:
- en
---
# OpenHermes-7B
## Model description
OpenHermes 7B is the first fine tune of the Hermes dataset that has a fully open source dataset!
What is unique about this 7B model is that it used sample packing, which speeds up training by many multiples if the dataset token averages arent near the max sequence length.
OpenHermes was trained on 242,000 entries of primarily GPT-4 generated data, from open datasets across the AI landscape, including:
- GPTeacher - General Instruct, Roleplay v1, Roleplay v2, and Code Instruct Datasets, by Teknium
- WizardLM (v1, evol_instruct 70k), by WizardLM Team/nlpxucan
- Airoboros GPT-4 (v1.0), by JonDurbin
- Camel-AI's domain expert datasets, by the Camel-AI Team
- CodeAlpaca, by Sahil2801
- GPT4-LLM and Unnatural Instructions, by Microsoft
Filtering included removal of OpenAI refusals, disclaimers, and "As an AI" type examples and more
The base dataset mix the model was trained on is identical to Nous-Hermes', minus the Nous-Instruct and PDACTL datasets which were private datasets.
The WANDB Project is public and can be examined at this link: https://wandb.ai/teknium1/openhermes/runs/openhermes-v2-qlora-7b-packed
Huge thank you to [main_horse](https://twitter.com/main_horse) for compute access and a16z for sponsoring my work, and all the dataset creators and other people who's work has contributed to this project!
## Benchmark Information
## Benchmark Results
GPT-4All Benchmark Set
```
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.4727|± |0.0146|
| | |acc_norm|0.4957|± |0.0146|
|arc_easy | 0|acc |0.7862|± |0.0084|
| | |acc_norm|0.7643|± |0.0087|
|boolq | 1|acc |0.7801|± |0.0072|
|hellaswag | 0|acc |0.5789|± |0.0049|
| | |acc_norm|0.7654|± |0.0042|
|openbookqa | 0|acc |0.3480|± |0.0213|
| | |acc_norm|0.4500|± |0.0223|
|piqa | 0|acc |0.7867|± |0.0096|
| | |acc_norm|0.7938|± |0.0094|
|winogrande | 0|acc |0.7048|± |0.0128|
Average: 0.679
```
BigBench:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------------------------|------:|---------------------|-----:|---|-----:|
|bigbench_causal_judgement | 0|multiple_choice_grade|0.5000|± |0.0364|
|bigbench_date_understanding | 0|multiple_choice_grade|0.5908|± |0.0256|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|0.3023|± |0.0286|
|bigbench_geometric_shapes | 0|multiple_choice_grade|0.1003|± |0.0159|
| | |exact_str_match |0.0000|± |0.0000|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|0.2520|± |0.0194|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|0.1871|± |0.0148|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|0.3833|± |0.0281|
|bigbench_movie_recommendation | 0|multiple_choice_grade|0.2500|± |0.0194|
|bigbench_navigate | 0|multiple_choice_grade|0.5000|± |0.0158|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|0.4370|± |0.0111|
|bigbench_ruin_names | 0|multiple_choice_grade|0.2679|± |0.0209|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|0.2495|± |0.0137|
|bigbench_snarks | 0|multiple_choice_grade|0.5249|± |0.0372|
|bigbench_sports_understanding | 0|multiple_choice_grade|0.5406|± |0.0159|
|bigbench_temporal_sequences | 0|multiple_choice_grade|0.2470|± |0.0136|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|0.1944|± |0.0112|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|0.1509|± |0.0086|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|0.3833|± |0.0281|
Average: 0.3367
```
AGI Eval
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------|------:|--------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.2441|± |0.0270|
| | |acc_norm|0.2402|± |0.0269|
|agieval_logiqa_en | 0|acc |0.2458|± |0.0169|
| | |acc_norm|0.2965|± |0.0179|
|agieval_lsat_ar | 0|acc |0.2522|± |0.0287|
| | |acc_norm|0.2130|± |0.0271|
|agieval_lsat_lr | 0|acc |0.2745|± |0.0198|
| | |acc_norm|0.2686|± |0.0196|
|agieval_lsat_rc | 0|acc |0.2900|± |0.0277|
| | |acc_norm|0.2379|± |0.0260|
|agieval_sat_en | 0|acc |0.4466|± |0.0347|
| | |acc_norm|0.3738|± |0.0338|
|agieval_sat_en_without_passage| 0|acc |0.3738|± |0.0338|
| | |acc_norm|0.3301|± |0.0328|
|agieval_sat_math | 0|acc |0.2318|± |0.0285|
| | |acc_norm|0.1864|± |0.0263|
Average: 0.2683
```
TruthfulQA:
```
hf-causal-experimental (pretrained=teknium/OpenHermes-7B,dtype=float16), limit: None, provide_description: False, num_fewshot: 0, batch_size: 8
| Task |Version|Metric|Value | |Stderr|
|-------------|------:|------|-----:|---|-----:|
|truthfulqa_mc| 1|mc2 |0.4542|± |0.0148|
```
## Training procedure
| 6,373 | [
[
-0.047149658203125,
-0.04107666015625,
0.0165557861328125,
0.006378173828125,
-0.01424407958984375,
-0.00276947021484375,
-0.00887298583984375,
-0.029388427734375,
0.034149169921875,
0.0079803466796875,
-0.040283203125,
-0.0565185546875,
-0.05548095703125,
-0.001865386962890625,
0.00838470458984375,
0.06683349609375,
-0.0033893585205078125,
-0.0019664764404296875,
0.01873779296875,
-0.027496337890625,
-0.037353515625,
-0.0003707408905029297,
-0.05596923828125,
-0.00972747802734375,
0.024566650390625,
0.033966064453125,
0.05078125,
0.048095703125,
0.051849365234375,
0.02362060546875,
-0.01800537109375,
0.00249481201171875,
-0.0252838134765625,
-0.0197906494140625,
0.0029468536376953125,
-0.031280517578125,
-0.0452880859375,
0.00885009765625,
0.044158935546875,
0.0474853515625,
-0.0062713623046875,
0.040740966796875,
0.00864410400390625,
0.0701904296875,
-0.036224365234375,
0.0229034423828125,
-0.004913330078125,
0.0004355907440185547,
-0.020294189453125,
-0.01143646240234375,
-0.0005717277526855469,
-0.032867431640625,
-0.0103759765625,
-0.052825927734375,
0.0161285400390625,
0.0113372802734375,
0.09515380859375,
0.0204315185546875,
-0.018341064453125,
-0.0139617919921875,
-0.033203125,
0.060546875,
-0.0552978515625,
0.01348114013671875,
0.034820556640625,
0.01273345947265625,
-0.0146636962890625,
-0.033721923828125,
-0.049896240234375,
0.00449371337890625,
-0.019683837890625,
0.030242919921875,
-0.01328277587890625,
-0.01910400390625,
0.030181884765625,
0.049835205078125,
-0.05511474609375,
0.00846099853515625,
-0.051544189453125,
-0.00811767578125,
0.053436279296875,
0.024627685546875,
0.01444244384765625,
-0.01000213623046875,
-0.029266357421875,
-0.031402587890625,
-0.0265045166015625,
0.038116455078125,
0.029083251953125,
0.01294708251953125,
-0.032745361328125,
0.05322265625,
-0.0283660888671875,
0.0352783203125,
0.0172882080078125,
-0.0157623291015625,
0.06292724609375,
-0.0283355712890625,
-0.023223876953125,
0.00667572021484375,
0.06536865234375,
0.04345703125,
-0.00846099853515625,
0.0158843994140625,
0.0101470947265625,
0.00870513916015625,
0.0009226799011230469,
-0.062744140625,
-0.011627197265625,
0.036224365234375,
-0.0406494140625,
-0.00852203369140625,
0.019439697265625,
-0.06591796875,
0.00029778480529785156,
-0.0150299072265625,
0.02581787109375,
-0.044830322265625,
-0.0191802978515625,
0.003993988037109375,
-0.0119781494140625,
0.029693603515625,
0.0191802978515625,
-0.04644775390625,
0.00937652587890625,
0.021209716796875,
0.06298828125,
-0.00707244873046875,
-0.0135345458984375,
-0.0176544189453125,
0.016326904296875,
-0.042022705078125,
0.045440673828125,
-0.01337432861328125,
-0.023345947265625,
-0.026824951171875,
0.022796630859375,
-0.01146697998046875,
-0.02374267578125,
0.051513671875,
-0.0113525390625,
0.0269317626953125,
-0.035125732421875,
-0.02984619140625,
-0.02069091796875,
0.0245819091796875,
-0.060028076171875,
0.0987548828125,
0.01558685302734375,
-0.0728759765625,
0.05413818359375,
-0.059295654296875,
0.00604248046875,
-0.01029205322265625,
-0.01007843017578125,
-0.055694580078125,
-0.01904296875,
0.027496337890625,
0.025299072265625,
-0.03546142578125,
0.00849151611328125,
-0.0233612060546875,
-0.02764892578125,
-0.004581451416015625,
-0.01468658447265625,
0.0841064453125,
0.0210418701171875,
-0.0509033203125,
0.011749267578125,
-0.0762939453125,
0.012359619140625,
0.02850341796875,
-0.0294952392578125,
-0.00811767578125,
-0.0206756591796875,
-0.02410888671875,
0.024566650390625,
0.0116424560546875,
-0.04376220703125,
0.018951416015625,
-0.02886962890625,
0.009735107421875,
0.06134033203125,
0.004100799560546875,
0.0167999267578125,
-0.05072021484375,
0.026031494140625,
0.0214385986328125,
0.00926971435546875,
0.007648468017578125,
-0.04217529296875,
-0.050567626953125,
-0.05145263671875,
0.00730133056640625,
0.038604736328125,
-0.022491455078125,
0.0506591796875,
-0.016693115234375,
-0.05535888671875,
-0.0474853515625,
-0.0074920654296875,
0.03070068359375,
0.0430908203125,
0.0419921875,
-0.020111083984375,
-0.030487060546875,
-0.07269287109375,
-0.00518035888671875,
-0.007526397705078125,
0.0011243820190429688,
0.02557373046875,
0.06622314453125,
-0.0013790130615234375,
0.056182861328125,
-0.05548095703125,
-0.04156494140625,
-0.0156097412109375,
-0.00778961181640625,
0.050567626953125,
0.05706787109375,
0.053985595703125,
-0.041229248046875,
-0.049835205078125,
-0.002460479736328125,
-0.06353759765625,
0.006961822509765625,
0.01105499267578125,
-0.0246734619140625,
0.0217742919921875,
0.024261474609375,
-0.0538330078125,
0.065673828125,
0.025970458984375,
-0.044586181640625,
0.058746337890625,
-0.0290985107421875,
0.030853271484375,
-0.07904052734375,
0.036773681640625,
-0.0003902912139892578,
0.0143890380859375,
-0.02349853515625,
-0.01104736328125,
0.003925323486328125,
0.007965087890625,
-0.0308074951171875,
0.05963134765625,
-0.0496826171875,
-0.0006780624389648438,
0.022369384765625,
-0.00677490234375,
-0.01290130615234375,
0.050567626953125,
-0.0034809112548828125,
0.0635986328125,
0.05548095703125,
-0.0364990234375,
0.0103759765625,
0.024871826171875,
-0.039520263671875,
0.04388427734375,
-0.0487060546875,
-0.0127105712890625,
-0.01183319091796875,
0.0145721435546875,
-0.08258056640625,
-0.0270538330078125,
0.0306396484375,
-0.0426025390625,
0.0205230712890625,
0.004070281982421875,
-0.0186767578125,
-0.0634765625,
-0.0472412109375,
0.012908935546875,
0.02978515625,
-0.0264129638671875,
0.0136566162109375,
0.005519866943359375,
-0.00015461444854736328,
-0.052520751953125,
-0.0458984375,
-0.027862548828125,
-0.0180206298828125,
-0.044647216796875,
0.02679443359375,
-0.01395416259765625,
-0.0093994140625,
0.0147857666015625,
-0.0234375,
-0.00439453125,
-0.001102447509765625,
0.02447509765625,
0.033050537109375,
-0.023406982421875,
-0.0167999267578125,
-0.011444091796875,
-0.016632080078125,
-0.0038166046142578125,
0.0047149658203125,
0.036041259765625,
-0.012451171875,
-0.03118896484375,
-0.04876708984375,
0.0172576904296875,
0.04638671875,
-0.0245361328125,
0.0870361328125,
0.04962158203125,
-0.0143890380859375,
0.0087890625,
-0.0269622802734375,
-0.0029010772705078125,
-0.033538818359375,
0.015960693359375,
-0.037322998046875,
-0.05963134765625,
0.044158935546875,
0.0136260986328125,
0.0188751220703125,
0.06439208984375,
0.045867919921875,
-0.0068511962890625,
0.07659912109375,
0.0204315185546875,
-0.007801055908203125,
0.0238189697265625,
-0.05487060546875,
0.00514984130859375,
-0.059478759765625,
-0.031951904296875,
-0.0491943359375,
-0.03900146484375,
-0.045806884765625,
-0.02203369140625,
0.02532958984375,
0.00601959228515625,
-0.056640625,
0.009552001953125,
-0.0506591796875,
0.0202178955078125,
0.06011962890625,
0.0299224853515625,
0.006931304931640625,
-0.0048675537109375,
-0.026153564453125,
-0.0025177001953125,
-0.041259765625,
-0.03472900390625,
0.09881591796875,
0.00975799560546875,
0.038726806640625,
0.0214691162109375,
0.051361083984375,
0.021331787109375,
0.0084686279296875,
-0.02972412109375,
0.03363037109375,
0.00878143310546875,
-0.060638427734375,
-0.030242919921875,
-0.02789306640625,
-0.0919189453125,
0.032379150390625,
-0.0257110595703125,
-0.06707763671875,
0.0206451416015625,
0.0142364501953125,
-0.0220794677734375,
0.03790283203125,
-0.046722412109375,
0.0718994140625,
-0.00345611572265625,
-0.0435791015625,
-0.007415771484375,
-0.0538330078125,
0.026519775390625,
0.00698089599609375,
0.030853271484375,
-0.020751953125,
0.005191802978515625,
0.0670166015625,
-0.054656982421875,
0.033966064453125,
-0.014801025390625,
0.01715087890625,
0.033905029296875,
-0.00807952880859375,
0.041351318359375,
-0.0020656585693359375,
-0.007160186767578125,
0.0085601806640625,
0.004039764404296875,
-0.06097412109375,
-0.00594329833984375,
0.05120849609375,
-0.0709228515625,
-0.04974365234375,
-0.057830810546875,
-0.03448486328125,
0.00007814168930053711,
0.032989501953125,
0.025115966796875,
0.0240325927734375,
-0.005565643310546875,
0.0278472900390625,
0.03851318359375,
-0.0257415771484375,
0.040130615234375,
0.0239410400390625,
0.004711151123046875,
-0.036468505859375,
0.059234619140625,
0.003925323486328125,
0.0170745849609375,
0.005035400390625,
0.0197296142578125,
-0.0191650390625,
-0.034271240234375,
-0.029632568359375,
0.0303955078125,
-0.0178985595703125,
-0.0175018310546875,
-0.03802490234375,
-0.0025005340576171875,
-0.04913330078125,
-0.02490234375,
-0.021514892578125,
-0.02728271484375,
-0.01654052734375,
-0.0156097412109375,
0.0301666259765625,
0.04034423828125,
-0.0210418701171875,
0.01068878173828125,
-0.04791259765625,
0.0309906005859375,
0.007080078125,
0.0293121337890625,
-0.0063323974609375,
-0.039306640625,
-0.017547607421875,
0.006866455078125,
-0.037384033203125,
-0.06829833984375,
0.04937744140625,
-0.005352020263671875,
0.05224609375,
0.02728271484375,
-0.003108978271484375,
0.058074951171875,
-0.0040130615234375,
0.07318115234375,
0.025177001953125,
-0.0411376953125,
0.04840087890625,
-0.0269927978515625,
0.033172607421875,
0.0455322265625,
0.042236328125,
-0.021942138671875,
-0.032440185546875,
-0.06640625,
-0.07073974609375,
0.08758544921875,
0.0288848876953125,
-0.03729248046875,
0.0135498046875,
0.004665374755859375,
-0.0093231201171875,
0.0129852294921875,
-0.04766845703125,
-0.05743408203125,
-0.0065155029296875,
-0.02392578125,
-0.01448822021484375,
-0.00070953369140625,
-0.0178680419921875,
-0.0396728515625,
0.05450439453125,
0.0096893310546875,
0.0360107421875,
0.019683837890625,
0.00348663330078125,
0.0011701583862304688,
0.01442718505859375,
0.034759521484375,
0.04351806640625,
-0.032318115234375,
-0.00624847412109375,
0.01215362548828125,
-0.055267333984375,
0.0128631591796875,
0.0037078857421875,
-0.0150604248046875,
-0.0106964111328125,
0.0270538330078125,
0.0469970703125,
-0.00998687744140625,
-0.0252838134765625,
0.0352783203125,
0.007843017578125,
-0.039337158203125,
-0.0229949951171875,
0.00994873046875,
-0.00968170166015625,
0.0279083251953125,
0.0240478515625,
0.007686614990234375,
0.00311279296875,
-0.04437255859375,
0.0026378631591796875,
0.0255126953125,
-0.0236663818359375,
-0.006439208984375,
0.06170654296875,
-0.011138916015625,
-0.0020999908447265625,
0.03338623046875,
-0.0120697021484375,
-0.033111572265625,
0.07244873046875,
0.01953125,
0.039459228515625,
-0.0283355712890625,
0.00652313232421875,
0.08056640625,
0.0311737060546875,
-0.00882720947265625,
0.037322998046875,
0.0197296142578125,
-0.0218505859375,
-0.00183868408203125,
-0.0501708984375,
-0.0106201171875,
0.02374267578125,
-0.0516357421875,
0.0214385986328125,
-0.02593994140625,
-0.0161895751953125,
0.0016307830810546875,
0.024139404296875,
-0.06622314453125,
0.0396728515625,
-0.01015472412109375,
0.06829833984375,
-0.0609130859375,
0.0518798828125,
0.0570068359375,
-0.0550537109375,
-0.08331298828125,
-0.024200439453125,
0.001949310302734375,
-0.052581787109375,
0.0462646484375,
0.00997161865234375,
0.025146484375,
-0.004329681396484375,
-0.031768798828125,
-0.094482421875,
0.1048583984375,
0.00518798828125,
-0.0295562744140625,
0.01030731201171875,
0.013031005859375,
0.031585693359375,
0.00887298583984375,
0.04266357421875,
0.04119873046875,
0.05328369140625,
0.0061798095703125,
-0.06787109375,
0.02325439453125,
-0.038055419921875,
-0.019439697265625,
0.03106689453125,
-0.07635498046875,
0.0804443359375,
-0.0169830322265625,
0.00099945068359375,
-0.02093505859375,
0.042694091796875,
0.03662109375,
0.024932861328125,
0.030487060546875,
0.06951904296875,
0.06121826171875,
-0.0181884765625,
0.07684326171875,
-0.02520751953125,
0.038177490234375,
0.07025146484375,
0.0177764892578125,
0.04693603515625,
0.026397705078125,
-0.0450439453125,
0.0268707275390625,
0.056060791015625,
-0.006961822509765625,
0.0291748046875,
0.0013036727905273438,
-0.0018167495727539062,
0.0003185272216796875,
0.0269927978515625,
-0.052734375,
0.012908935546875,
0.0154266357421875,
-0.0195159912109375,
-0.005855560302734375,
-0.01203155517578125,
0.0176544189453125,
-0.0104217529296875,
-0.030029296875,
0.038818359375,
-0.014556884765625,
-0.0482177734375,
0.053558349609375,
-0.007781982421875,
0.0423583984375,
-0.048828125,
0.0009551048278808594,
-0.0158538818359375,
0.0264892578125,
-0.035003662109375,
-0.0787353515625,
0.0204620361328125,
-0.001949310302734375,
-0.0096588134765625,
0.00464630126953125,
0.0270233154296875,
-0.01013946533203125,
-0.0384521484375,
0.01544952392578125,
0.025543212890625,
0.0171966552734375,
0.0107879638671875,
-0.059967041015625,
-0.004528045654296875,
0.0125579833984375,
-0.05413818359375,
0.022003173828125,
0.039703369140625,
-0.0032863616943359375,
0.035797119140625,
0.059112548828125,
0.00594329833984375,
0.0192718505859375,
-0.032012939453125,
0.0711669921875,
-0.062744140625,
-0.042083740234375,
-0.051513671875,
0.042236328125,
-0.022247314453125,
-0.046142578125,
0.06787109375,
0.0650634765625,
0.05078125,
-0.000629425048828125,
0.04931640625,
-0.03851318359375,
0.045654296875,
-0.0297088623046875,
0.044891357421875,
-0.062744140625,
-0.0115509033203125,
-0.024749755859375,
-0.055633544921875,
-0.0164947509765625,
0.054412841796875,
-0.035400390625,
0.00897216796875,
0.06207275390625,
0.0584716796875,
0.0015935897827148438,
0.0014772415161132812,
-0.0023708343505859375,
0.029327392578125,
0.0184326171875,
0.05487060546875,
0.021759033203125,
-0.033477783203125,
0.045684814453125,
-0.034332275390625,
-0.023284912109375,
-0.0257568359375,
-0.03936767578125,
-0.056854248046875,
-0.040771484375,
-0.030853271484375,
-0.037994384765625,
-0.00806427001953125,
0.06903076171875,
0.03253173828125,
-0.0528564453125,
-0.0213775634765625,
-0.0004382133483886719,
-0.0026073455810546875,
-0.03448486328125,
-0.01605224609375,
0.063720703125,
-0.01071929931640625,
-0.0545654296875,
-0.0017547607421875,
-0.00582122802734375,
0.00814056396484375,
0.01312255859375,
-0.01264190673828125,
-0.0308685302734375,
0.006702423095703125,
0.02911376953125,
0.0272216796875,
-0.034820556640625,
-0.01526641845703125,
-0.0108184814453125,
-0.023101806640625,
0.034820556640625,
0.00502777099609375,
-0.0303192138671875,
0.01537322998046875,
0.03326416015625,
0.0186004638671875,
0.0625,
-0.0010309219360351562,
-0.0023899078369140625,
-0.0174102783203125,
0.0165863037109375,
-0.001323699951171875,
0.02691650390625,
-0.00444793701171875,
-0.0180206298828125,
0.05804443359375,
0.0245208740234375,
-0.043182373046875,
-0.06475830078125,
-0.0240020751953125,
-0.09814453125,
-0.0211029052734375,
0.0770263671875,
-0.010162353515625,
-0.04217529296875,
0.00730133056640625,
-0.035308837890625,
0.0206146240234375,
-0.047698974609375,
0.04229736328125,
0.044921875,
-0.0147247314453125,
0.0116424560546875,
-0.06121826171875,
0.0277099609375,
0.023223876953125,
-0.061370849609375,
-0.0158843994140625,
0.0389404296875,
0.024688720703125,
0.0103759765625,
0.05743408203125,
-0.01568603515625,
0.0168914794921875,
0.02325439453125,
0.0168609619140625,
-0.00714874267578125,
0.005466461181640625,
0.00200653076171875,
0.029296875,
-0.009918212890625,
-0.045654296875
]
] |
openbmb/UltraLM-13b | 2023-06-27T09:20:53.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:stingning/ultrachat",
"arxiv:2305.14233",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | openbmb | null | null | openbmb/UltraLM-13b | 69 | 7,528 | transformers | 2023-06-26T06:43:47 | ---
datasets:
- stingning/ultrachat
---
# UltraLM-13b
<!-- Provide a quick summary of what the model is/does. -->
This is UltraLM-13b delta weights, a chat language model trained upon [UltraChat](https://github.com/thunlp/UltraChat)
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
The model is fine-tuned based on LLaMA-13b with a multi-turn chat-format template as below
```
User: instruction 1<eos_token>
Assistant: response 1<eos_token>
User: instruction 2<eos_token>
Assistant: response 2<eos_token>
...
```
- **License:** UltraLM is based on LLaMA and should be used under LLaMA's [model license](https://github.com/facebookresearch/llama/blob/main/MODEL_CARD.md).
- **Finetuned from model:** LLaMA-13b
- **Finetuned on data:** [UltraChat](https://github.com/thunlp/UltraChat)
### Model Sources
<!-- Provide the basic links for the model. -->
- **Repository:** [UltraChat](https://github.com/thunlp/UltraChat)
- **Paper:** [arxiv](https://arxiv.org/abs/2305.14233)
- **Demo:** [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. -->
To use this model, you need to [recover](https://github.com/thunlp/UltraChat/tree/main/UltraLM) the full model from the delta weights and perform inference following the template below:
```
[Optional]User: system prompt<eos_token>
User: user input<eos_token>
Assistant:
```
| 1,504 | [
[
-0.0037517547607421875,
-0.058013916015625,
0.0228424072265625,
0.02813720703125,
-0.044403076171875,
0.0130615234375,
0.006305694580078125,
-0.033050537109375,
0.033416748046875,
0.049591064453125,
-0.0604248046875,
-0.0300445556640625,
-0.027557373046875,
-0.01367950439453125,
-0.03302001953125,
0.0745849609375,
0.0252227783203125,
0.01480865478515625,
0.01861572265625,
0.0013151168823242188,
-0.032684326171875,
-0.052215576171875,
-0.044525146484375,
-0.030487060546875,
0.01349639892578125,
0.03411865234375,
0.042022705078125,
0.034332275390625,
0.025909423828125,
0.0187530517578125,
-0.03594970703125,
0.01143646240234375,
-0.057952880859375,
-0.0027294158935546875,
0.025787353515625,
-0.0207061767578125,
-0.058441162109375,
-0.026702880859375,
0.045623779296875,
0.029388427734375,
-0.040557861328125,
0.0201568603515625,
0.0045623779296875,
0.0249481201171875,
-0.03387451171875,
0.00937652587890625,
-0.039886474609375,
-0.01519012451171875,
-0.0194244384765625,
0.002605438232421875,
-0.03387451171875,
0.0003445148468017578,
-0.00963592529296875,
-0.037689208984375,
-0.0221099853515625,
0.0141143798828125,
0.0859375,
0.00890350341796875,
-0.045623779296875,
-0.0094146728515625,
-0.061920166015625,
0.045806884765625,
-0.080078125,
0.038970947265625,
0.0303497314453125,
0.0265655517578125,
-0.01548004150390625,
-0.063232421875,
-0.038116455078125,
-0.0246429443359375,
0.0174407958984375,
-0.0012693405151367188,
-0.030303955078125,
0.00734710693359375,
0.008453369140625,
0.0174713134765625,
-0.0182952880859375,
0.0176544189453125,
-0.039337158203125,
-0.0249786376953125,
0.0362548828125,
0.039337158203125,
0.002292633056640625,
-0.0168304443359375,
-0.036163330078125,
-0.00894927978515625,
-0.043212890625,
0.0003960132598876953,
0.023101806640625,
0.03961181640625,
-0.048370361328125,
0.05413818359375,
-0.0237579345703125,
0.0494384765625,
0.005901336669921875,
-0.00908660888671875,
0.0202789306640625,
-0.0199127197265625,
-0.03961181640625,
-0.01200103759765625,
0.061309814453125,
0.03814697265625,
0.004337310791015625,
0.026092529296875,
-0.021148681640625,
-0.03179931640625,
0.003208160400390625,
-0.053192138671875,
-0.008575439453125,
0.009796142578125,
-0.050048828125,
-0.025177001953125,
-0.0091552734375,
-0.040435791015625,
-0.036376953125,
-0.00858306884765625,
0.01873779296875,
-0.022064208984375,
-0.032196044921875,
0.0012903213500976562,
0.0184783935546875,
0.021331787109375,
0.015411376953125,
-0.0469970703125,
0.02191162109375,
0.0191192626953125,
0.053985595703125,
0.01544189453125,
-0.011810302734375,
0.0010442733764648438,
0.00792694091796875,
-0.0252685546875,
0.0523681640625,
-0.0271148681640625,
-0.037384033203125,
-0.002880096435546875,
0.01010894775390625,
0.010528564453125,
-0.0190887451171875,
0.061309814453125,
-0.051025390625,
0.005950927734375,
-0.020111083984375,
-0.0477294921875,
-0.020416259765625,
-0.0027790069580078125,
-0.0394287109375,
0.09320068359375,
-0.005527496337890625,
-0.05181884765625,
-0.0013074874877929688,
-0.052520751953125,
-0.00605010986328125,
0.022979736328125,
0.0163116455078125,
-0.0145111083984375,
0.00511932373046875,
-0.0010957717895507812,
0.0439453125,
-0.028717041015625,
0.0145263671875,
-0.04205322265625,
-0.04107666015625,
0.01261138916015625,
-0.028076171875,
0.08734130859375,
0.02496337890625,
-0.0084991455078125,
0.02056884765625,
-0.06463623046875,
-0.0091552734375,
0.0279693603515625,
-0.038818359375,
-0.0322265625,
-0.0152587890625,
0.024627685546875,
0.002819061279296875,
0.038909912109375,
-0.0458984375,
0.0196685791015625,
-0.0009984970092773438,
0.01474761962890625,
0.07135009765625,
-0.00679779052734375,
0.0152435302734375,
-0.005741119384765625,
0.028289794921875,
-0.006011962890625,
0.03515625,
-0.0017299652099609375,
-0.039886474609375,
-0.06427001953125,
-0.03302001953125,
0.0117340087890625,
0.04754638671875,
-0.043304443359375,
0.053466796875,
-0.01003265380859375,
-0.05975341796875,
-0.044647216796875,
0.007617950439453125,
0.0303192138671875,
0.034515380859375,
0.0193634033203125,
-0.0163726806640625,
-0.0660400390625,
-0.06787109375,
0.01145172119140625,
-0.032196044921875,
0.0017442703247070312,
0.031524658203125,
0.035888671875,
-0.038848876953125,
0.045928955078125,
-0.040191650390625,
-0.0110015869140625,
-0.0283966064453125,
0.002101898193359375,
0.00872802734375,
0.0278778076171875,
0.035308837890625,
-0.031158447265625,
-0.02410888671875,
-0.022979736328125,
-0.048126220703125,
-0.0166778564453125,
-0.01297760009765625,
-0.02569580078125,
0.0226287841796875,
0.02984619140625,
-0.039093017578125,
0.03814697265625,
0.04827880859375,
-0.042266845703125,
0.044525146484375,
-0.0078582763671875,
-0.022369384765625,
-0.0941162109375,
-0.004604339599609375,
-0.0089874267578125,
-0.01302337646484375,
-0.044219970703125,
0.0016355514526367188,
0.01290130615234375,
0.0186920166015625,
-0.040771484375,
0.061004638671875,
-0.032501220703125,
-0.0079193115234375,
-0.03363037109375,
-0.0299224853515625,
-0.0177154541015625,
0.041595458984375,
-0.01447296142578125,
0.035308837890625,
0.0460205078125,
-0.048858642578125,
0.052459716796875,
0.01617431640625,
-0.007904052734375,
0.032989501953125,
-0.060150146484375,
0.027069091796875,
0.01412200927734375,
0.040130615234375,
-0.09136962890625,
-0.01215362548828125,
0.0221710205078125,
-0.035003662109375,
0.01071929931640625,
-0.0127410888671875,
-0.04193115234375,
-0.038421630859375,
-0.017333984375,
0.0014190673828125,
0.037506103515625,
-0.029327392578125,
0.051177978515625,
0.03399658203125,
0.0095672607421875,
-0.050140380859375,
-0.058685302734375,
0.01253509521484375,
-0.01241302490234375,
-0.0203094482421875,
0.00225067138671875,
-0.00911712646484375,
-0.004779815673828125,
-0.005710601806640625,
-0.002475738525390625,
-0.014068603515625,
0.0057525634765625,
0.02880859375,
0.036651611328125,
0.00022494792938232422,
-0.0079193115234375,
0.0038433074951171875,
0.0024967193603515625,
-0.0135345458984375,
-0.0032024383544921875,
0.07159423828125,
-0.018280029296875,
-0.0254058837890625,
-0.06439208984375,
0.00681304931640625,
0.042266845703125,
0.00749969482421875,
0.07574462890625,
0.041748046875,
-0.0242919921875,
0.0145416259765625,
-0.052490234375,
-0.024810791015625,
-0.03546142578125,
0.03375244140625,
-0.020538330078125,
-0.058013916015625,
0.0614013671875,
0.0161590576171875,
0.0107269287109375,
0.0308685302734375,
0.057952880859375,
-0.019287109375,
0.06634521484375,
0.07501220703125,
-0.006496429443359375,
0.057281494140625,
-0.01605224609375,
-0.01142120361328125,
-0.0789794921875,
-0.0307769775390625,
-0.038238525390625,
-0.013214111328125,
-0.0338134765625,
-0.01145172119140625,
-0.003631591796875,
0.018951416015625,
-0.0484619140625,
0.041900634765625,
-0.0306396484375,
0.0172576904296875,
0.056732177734375,
0.003536224365234375,
0.0243072509765625,
0.00475311279296875,
-0.03082275390625,
0.0084075927734375,
-0.043304443359375,
-0.05596923828125,
0.0992431640625,
0.031280517578125,
0.0850830078125,
0.005397796630859375,
0.043365478515625,
0.01409912109375,
0.0205535888671875,
-0.03656005859375,
0.05267333984375,
0.008636474609375,
-0.07000732421875,
-0.0155181884765625,
-0.0391845703125,
-0.06732177734375,
0.0167083740234375,
-0.01161956787109375,
-0.059906005859375,
0.00783538818359375,
0.0013895034790039062,
-0.0179443359375,
0.03265380859375,
-0.050323486328125,
0.036712646484375,
-0.0240325927734375,
0.01824951171875,
-0.00714111328125,
-0.046051025390625,
0.04644775390625,
-0.0050048828125,
0.01947021484375,
-0.00963592529296875,
-0.0033321380615234375,
0.06988525390625,
-0.0287933349609375,
0.09600830078125,
-0.01177215576171875,
-0.0157928466796875,
0.019989013671875,
0.0177154541015625,
0.0302276611328125,
-0.005313873291015625,
-0.0019512176513671875,
0.031585693359375,
0.006412506103515625,
-0.0305633544921875,
-0.0068206787109375,
0.045806884765625,
-0.086181640625,
-0.01438140869140625,
-0.017120361328125,
-0.03790283203125,
0.0016641616821289062,
0.022491455078125,
0.01523590087890625,
0.023895263671875,
-0.00139617919921875,
0.02191162109375,
0.022705078125,
-0.01629638671875,
0.03338623046875,
0.06646728515625,
-0.0311737060546875,
-0.0236053466796875,
0.034942626953125,
-0.003681182861328125,
0.0127716064453125,
0.019927978515625,
-0.003772735595703125,
-0.034515380859375,
-0.0182952880859375,
-0.03607177734375,
0.0227203369140625,
-0.052886962890625,
-0.016876220703125,
-0.038116455078125,
-0.02410888671875,
-0.036773681640625,
0.014251708984375,
-0.0472412109375,
-0.03656005859375,
-0.03717041015625,
-0.01678466796875,
0.042877197265625,
0.056121826171875,
-0.01251983642578125,
0.06427001953125,
-0.061737060546875,
0.00951385498046875,
0.02276611328125,
0.020843505859375,
0.0206451416015625,
-0.07916259765625,
-0.0151519775390625,
0.0217742919921875,
-0.0340576171875,
-0.07080078125,
0.02288818359375,
0.0035495758056640625,
0.0546875,
0.03997802734375,
0.0111541748046875,
0.058258056640625,
-0.0216827392578125,
0.05511474609375,
0.01434326171875,
-0.0653076171875,
0.01934814453125,
-0.0135650634765625,
0.006134033203125,
0.01898193359375,
0.0220947265625,
-0.055572509765625,
-0.020233154296875,
-0.055938720703125,
-0.0638427734375,
0.057159423828125,
0.0182952880859375,
0.049835205078125,
0.01537322998046875,
0.0279083251953125,
-0.0179901123046875,
0.011932373046875,
-0.07965087890625,
-0.040435791015625,
-0.0144805908203125,
-0.01629638671875,
0.017120361328125,
-0.048370361328125,
-0.0121917724609375,
-0.0175018310546875,
0.05377197265625,
0.0019178390502929688,
0.02685546875,
-0.018341064453125,
0.004810333251953125,
-0.0211944580078125,
0.0033206939697265625,
0.044952392578125,
0.041412353515625,
-0.032379150390625,
-0.011962890625,
0.024017333984375,
-0.040283203125,
0.00341033935546875,
0.0152587890625,
-0.00972747802734375,
-0.01348876953125,
0.045684814453125,
0.0772705078125,
0.01045989990234375,
-0.03826904296875,
0.0199432373046875,
-0.020477294921875,
0.0026760101318359375,
-0.01105499267578125,
-0.02117919921875,
0.0328369140625,
0.0180816650390625,
0.049224853515625,
-0.028564453125,
-0.003932952880859375,
-0.023895263671875,
-0.0003535747528076172,
0.038238525390625,
-0.030853271484375,
-0.0242462158203125,
0.04302978515625,
0.031036376953125,
-0.028228759765625,
0.0478515625,
-0.0159149169921875,
-0.0273284912109375,
0.0360107421875,
0.042205810546875,
0.05157470703125,
-0.0300140380859375,
0.007061004638671875,
0.01117706298828125,
0.029296875,
0.00833892822265625,
0.04443359375,
0.00566864013671875,
-0.067138671875,
-0.032501220703125,
-0.053314208984375,
-0.037872314453125,
0.020843505859375,
-0.045623779296875,
0.026031494140625,
-0.034637451171875,
-0.02093505859375,
-0.013641357421875,
0.00698089599609375,
-0.0291290283203125,
-0.0011129379272460938,
0.0236358642578125,
0.07171630859375,
-0.060882568359375,
0.0712890625,
0.07421875,
-0.027496337890625,
-0.07659912109375,
-0.02081298828125,
0.0254364013671875,
-0.08935546875,
0.042449951171875,
-0.0122528076171875,
0.005161285400390625,
0.0044097900390625,
-0.07244873046875,
-0.0731201171875,
0.0799560546875,
0.03143310546875,
-0.050933837890625,
-0.0010576248168945312,
-0.0130615234375,
0.03668212890625,
-0.0221099853515625,
0.03955078125,
0.01270294189453125,
0.0161895751953125,
0.0382080078125,
-0.0980224609375,
0.007904052734375,
-0.005077362060546875,
0.0096893310546875,
-0.01288604736328125,
-0.059844970703125,
0.0771484375,
-0.010284423828125,
0.002414703369140625,
0.0164337158203125,
0.052520751953125,
0.047607421875,
-0.0018968582153320312,
0.038055419921875,
0.04180908203125,
0.0523681640625,
0.0133819580078125,
0.07086181640625,
-0.02886962890625,
0.053009033203125,
0.07952880859375,
-0.026947021484375,
0.056396484375,
0.0235137939453125,
-0.00914764404296875,
0.04449462890625,
0.0677490234375,
0.00521087646484375,
0.034423828125,
0.005550384521484375,
0.007625579833984375,
-0.0216827392578125,
-0.00983428955078125,
-0.038116455078125,
0.03656005859375,
0.0212860107421875,
-0.01406097412109375,
-0.016632080078125,
-0.01189422607421875,
0.0217742919921875,
-0.037750244140625,
-0.018829345703125,
0.053985595703125,
0.01271820068359375,
0.0004420280456542969,
0.07037353515625,
0.0196075439453125,
0.07183837890625,
-0.055572509765625,
-0.004913330078125,
-0.03790283203125,
0.0157470703125,
0.0006937980651855469,
-0.04541015625,
0.01401519775390625,
0.008544921875,
0.0132293701171875,
0.0007171630859375,
0.049072265625,
-0.00426483154296875,
-0.05731201171875,
0.039459228515625,
0.03863525390625,
0.031585693359375,
0.002056121826171875,
-0.0732421875,
0.0273590087890625,
0.014892578125,
-0.040740966796875,
0.0161895751953125,
0.0245361328125,
-0.0122222900390625,
0.0760498046875,
0.06390380859375,
0.0016870498657226562,
0.0056610107421875,
0.0122528076171875,
0.07781982421875,
-0.043731689453125,
-0.030609130859375,
-0.06573486328125,
0.0281982421875,
-0.0007138252258300781,
-0.035003662109375,
0.05352783203125,
0.03375244140625,
0.042694091796875,
0.00960540771484375,
0.0223846435546875,
-0.01617431640625,
0.035919189453125,
-0.0225067138671875,
0.0601806640625,
-0.056243896484375,
0.0225982666015625,
-0.0263824462890625,
-0.06396484375,
-0.0002505779266357422,
0.044158935546875,
0.017059326171875,
0.01535797119140625,
0.026763916015625,
0.05865478515625,
-0.0003325939178466797,
0.005123138427734375,
0.01873779296875,
0.038055419921875,
0.02435302734375,
0.031585693359375,
0.061981201171875,
-0.0260467529296875,
0.0625,
-0.0184783935546875,
-0.0340576171875,
-0.035400390625,
-0.062164306640625,
-0.0711669921875,
-0.043701171875,
-0.00115966796875,
-0.024169921875,
-0.0022792816162109375,
0.068115234375,
0.078369140625,
-0.045684814453125,
-0.048065185546875,
0.0169525146484375,
-0.00022590160369873047,
-0.00821685791015625,
-0.015045166015625,
0.01654052734375,
0.01019287109375,
-0.042633056640625,
0.02276611328125,
-0.01309967041015625,
0.0394287109375,
-0.0169525146484375,
-0.00960540771484375,
-0.0285186767578125,
0.0112457275390625,
0.029998779296875,
0.0218048095703125,
-0.059173583984375,
0.004985809326171875,
0.00423431396484375,
-0.024566650390625,
-0.0192718505859375,
0.0148468017578125,
-0.031463623046875,
0.00450897216796875,
0.00994110107421875,
0.0221099853515625,
0.04541015625,
-0.0010280609130859375,
0.025909423828125,
-0.05810546875,
0.0177764892578125,
0.0019779205322265625,
0.03466796875,
0.0303497314453125,
-0.0302276611328125,
0.032073974609375,
0.01226043701171875,
-0.0312042236328125,
-0.0821533203125,
0.0002913475036621094,
-0.07421875,
-0.02294921875,
0.0966796875,
0.00144195556640625,
-0.004852294921875,
0.0019435882568359375,
-0.042205810546875,
0.029937744140625,
-0.05316162109375,
0.046783447265625,
0.056060791015625,
-0.01059722900390625,
-0.0311431884765625,
-0.034942626953125,
0.01849365234375,
0.01739501953125,
-0.046417236328125,
0.0146331787109375,
0.03106689453125,
0.02197265625,
0.006465911865234375,
0.06719970703125,
0.00494384765625,
-0.002593994140625,
-0.0251617431640625,
0.04534912109375,
-0.01367950439453125,
-0.01540374755859375,
-0.0279693603515625,
-0.0234222412109375,
-0.005130767822265625,
-0.0185394287109375
]
] |
ItsJayQz/GTA5_Artwork_Diffusion | 2023-01-28T01:05:18.000Z | [
"diffusers",
"safetensors",
"stable-diffusion",
"text-to-image",
"grand theft auto",
"game",
"en",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | ItsJayQz | null | null | ItsJayQz/GTA5_Artwork_Diffusion | 105 | 7,520 | diffusers | 2022-12-13T03:04:55 | ---
language:
- en
license: creativeml-openrail-m
tags:
- stable-diffusion
- text-to-image
- diffusers
- grand theft auto
- game
inference: true
---
### GTA5 Artwork Diffusion
This model was trained on the loading screens, gta storymode, and gta online DLCs artworks.
Which includes characters, background, chop, and some objects.
The model can do people and portrait pretty easily, as well as cars, and houses.
For some reasons, the model stills automatically include in some game footage, so landscapes tend to look a bit more game-like.
Please check out important informations on the usage of the model down bellow.
To reference the art style, use the token: gtav style
There is already an existing model that uses textual inversion. This is trained using Dreambooth instead, whether or not this method is better, I will let you judge.
### Gradio
We support a [Gradio](https://github.com/gradio-app/gradio) Web UI to run GTA5_Artwork_Diffusion:
[](https://huggingface.co/spaces/ItsJayQz/GTA5_Artwork_Diffusion)
Here are some samples.
**Portraits**
Prompt used:
*name* in gtav style
Guidance: 7
Steps: 65 using DDIM
I'm not a prompt wizard so you can definitely get better results with some tuning.
**Landscapes**
**Objects**
**Disclaimers**
- I'm in no way affliated with Rockstar, or any entities relating to the ownership of the game artworks.
- The phrase GTA is simply a reference for accessibility.
- This was created entirely for research, and entertainment purpose.
- I did not plan, or is planning on turning this model into a commercial product, or use for commercial purposes.
- I do not condone the usage of the model for making counterfeit products that might infringe on Rockstar's copyrights/trademarks.
**License**
- This model is under Creative OpenRAIL-M.
- This means the model can be used royalty-free, and flexible with the model usage, such as redistribution of the model, or of any derivatives of the model.
- However, there are restrictions on the openess of the license.
More info into the restrictions can be found [here](https://huggingface.co/spaces/CompVis/stable-diffusion-license).
**Responsibilities**
- By using/downloading the model, you are responsible for:
- All outputs/usage of the model.
- Understanding the Disclaimers.
- Upholding the terms of the license.
Thanks for checking out the model! | 3,147 | [
[
-0.045196533203125,
-0.0675048828125,
0.06158447265625,
0.0305023193359375,
-0.022735595703125,
0.0243988037109375,
0.01641845703125,
-0.05828857421875,
0.0247039794921875,
0.044403076171875,
-0.046722412109375,
-0.05401611328125,
-0.03863525390625,
-0.0202789306640625,
-0.01393890380859375,
0.07470703125,
0.004688262939453125,
-0.01483154296875,
-0.00485992431640625,
0.0068359375,
-0.0489501953125,
-0.0006060600280761719,
-0.048614501953125,
-0.02764892578125,
0.02130126953125,
0.0007581710815429688,
0.07513427734375,
0.0535888671875,
0.018402099609375,
0.0196685791015625,
-0.022003173828125,
-0.002285003662109375,
-0.03363037109375,
-0.01496124267578125,
-0.0013751983642578125,
-0.007541656494140625,
-0.03802490234375,
0.003459930419921875,
0.005664825439453125,
-0.01824951171875,
-0.0243682861328125,
0.00954437255859375,
-0.02252197265625,
0.037933349609375,
-0.033843994140625,
0.011199951171875,
-0.0022068023681640625,
-0.0029850006103515625,
-0.01557159423828125,
0.0049591064453125,
-0.00304412841796875,
-0.0460205078125,
0.005046844482421875,
-0.0872802734375,
0.018096923828125,
-0.004665374755859375,
0.10845947265625,
0.002117156982421875,
-0.00853729248046875,
-0.024993896484375,
-0.062286376953125,
0.0294342041015625,
-0.060150146484375,
0.00238037109375,
-0.004741668701171875,
0.0343017578125,
0.00470733642578125,
-0.05291748046875,
-0.0362548828125,
-0.017181396484375,
0.005702972412109375,
0.0340576171875,
-0.02935791015625,
-0.0024585723876953125,
0.0151519775390625,
0.037628173828125,
-0.038726806640625,
-0.01465606689453125,
-0.0306549072265625,
-0.003925323486328125,
0.039093017578125,
0.01751708984375,
0.059112548828125,
-0.00911712646484375,
-0.053436279296875,
-0.0144805908203125,
-0.05926513671875,
0.0155792236328125,
0.035125732421875,
-0.0167388916015625,
-0.0335693359375,
0.0281524658203125,
0.0171051025390625,
0.0230865478515625,
0.0244140625,
-0.002353668212890625,
0.01788330078125,
-0.0212249755859375,
-0.0218505859375,
-0.05059814453125,
0.049560546875,
0.048553466796875,
-0.00417327880859375,
-0.0185699462890625,
-0.0105133056640625,
-0.0022754669189453125,
0.0225677490234375,
-0.06719970703125,
-0.034210205078125,
0.021209716796875,
-0.024078369140625,
-0.00006508827209472656,
-0.0065765380859375,
-0.08026123046875,
-0.0186004638671875,
-0.0016832351684570312,
0.039825439453125,
-0.02935791015625,
-0.029449462890625,
0.0247955322265625,
-0.03729248046875,
0.0145721435546875,
0.0261383056640625,
-0.04351806640625,
0.0032501220703125,
0.0195465087890625,
0.08270263671875,
0.01006317138671875,
0.00321197509765625,
0.0168304443359375,
0.03851318359375,
-0.015716552734375,
0.069580078125,
-0.02410888671875,
-0.0416259765625,
-0.015716552734375,
0.0033626556396484375,
0.0219879150390625,
-0.03863525390625,
0.031402587890625,
-0.045135498046875,
0.0105743408203125,
-0.034423828125,
-0.01355743408203125,
-0.04315185546875,
-0.01323699951171875,
-0.042938232421875,
0.067626953125,
0.01001739501953125,
-0.03363037109375,
0.03143310546875,
-0.08111572265625,
-0.0025691986083984375,
0.017486572265625,
-0.004299163818359375,
-0.041229248046875,
0.00327301025390625,
-0.00150299072265625,
0.0125732421875,
0.0016536712646484375,
0.01139068603515625,
-0.029754638671875,
-0.01288604736328125,
-0.013641357421875,
0.02764892578125,
0.0733642578125,
0.027618408203125,
-0.05877685546875,
0.0034999847412109375,
-0.032562255859375,
-0.005542755126953125,
0.0156707763671875,
-0.01434326171875,
-0.0269927978515625,
-0.00006443262100219727,
0.031402587890625,
0.046173095703125,
0.018280029296875,
-0.0274810791015625,
0.0287933349609375,
-0.01495361328125,
0.027557373046875,
0.06317138671875,
0.00948333740234375,
0.041351318359375,
-0.0231475830078125,
0.0565185546875,
0.01363372802734375,
0.017822265625,
0.028167724609375,
-0.056671142578125,
-0.03765869140625,
-0.04864501953125,
-0.00940704345703125,
0.032928466796875,
-0.05084228515625,
0.0227813720703125,
-0.0005102157592773438,
-0.049591064453125,
-0.01690673828125,
-0.006702423095703125,
0.0377197265625,
0.010711669921875,
0.00949859619140625,
-0.0482177734375,
-0.024993896484375,
-0.068115234375,
0.01056671142578125,
0.0025348663330078125,
-0.007564544677734375,
0.03839111328125,
0.050872802734375,
-0.0218658447265625,
0.0679931640625,
-0.059173583984375,
-0.02618408203125,
-0.0029163360595703125,
0.02630615234375,
-0.002735137939453125,
0.04290771484375,
0.0958251953125,
-0.041168212890625,
-0.0303802490234375,
-0.004161834716796875,
-0.051788330078125,
-0.01187896728515625,
-0.00418853759765625,
-0.03765869140625,
-0.02044677734375,
-0.0024051666259765625,
-0.07147216796875,
0.03619384765625,
0.0282745361328125,
-0.0521240234375,
0.03106689453125,
-0.04632568359375,
0.02606201171875,
-0.06915283203125,
0.008331298828125,
0.02874755859375,
-0.0248870849609375,
-0.044342041015625,
0.048828125,
-0.032135009765625,
-0.0101776123046875,
-0.0289459228515625,
0.0252532958984375,
-0.04901123046875,
-0.00249481201171875,
-0.032684326171875,
0.0011167526245117188,
-0.0005803108215332031,
0.0268096923828125,
0.01141357421875,
0.0753173828125,
0.05511474609375,
-0.04376220703125,
0.00876617431640625,
0.027008056640625,
-0.03582763671875,
0.03411865234375,
-0.060516357421875,
0.03497314453125,
-0.04010009765625,
0.026702880859375,
-0.0921630859375,
-0.006855010986328125,
0.053680419921875,
-0.02532958984375,
-0.010650634765625,
0.0005092620849609375,
-0.037872314453125,
-0.004184722900390625,
-0.00820159912109375,
0.01224517822265625,
0.05889892578125,
-0.021240234375,
0.06494140625,
0.04437255859375,
0.01898193359375,
-0.0244293212890625,
-0.06085205078125,
0.00412750244140625,
-0.054351806640625,
-0.06243896484375,
0.06036376953125,
-0.00933074951171875,
-0.02099609375,
-0.001483917236328125,
0.02288818359375,
-0.0291748046875,
-0.011016845703125,
0.049407958984375,
0.025146484375,
-0.004375457763671875,
-0.04449462890625,
-0.009552001953125,
0.0077056884765625,
0.0093231201171875,
0.0034618377685546875,
0.029815673828125,
0.003940582275390625,
0.007396697998046875,
-0.046600341796875,
0.0210723876953125,
0.0635986328125,
0.0065155029296875,
0.03424072265625,
0.039093017578125,
-0.02099609375,
0.0079345703125,
-0.02398681640625,
0.00078582763671875,
-0.033660888671875,
-0.0002751350402832031,
-0.0178680419921875,
-0.0299835205078125,
0.08447265625,
0.01009368896484375,
0.036346435546875,
0.044342041015625,
0.047576904296875,
-0.00485992431640625,
0.08026123046875,
0.058990478515625,
0.0116119384765625,
0.048187255859375,
-0.052764892578125,
-0.0019311904907226562,
-0.044677734375,
-0.044677734375,
-0.029144287109375,
-0.0369873046875,
-0.031280517578125,
-0.043060302734375,
0.016082763671875,
-0.01091766357421875,
-0.0267791748046875,
0.05450439453125,
-0.0309295654296875,
0.02618408203125,
0.0209808349609375,
0.045379638671875,
0.03192138671875,
0.0033550262451171875,
0.0116729736328125,
-0.01152801513671875,
-0.01265716552734375,
-0.040130615234375,
0.051116943359375,
0.0177154541015625,
0.051666259765625,
0.027801513671875,
0.03143310546875,
0.03350830078125,
0.0303802490234375,
-0.031402587890625,
0.0411376953125,
-0.0001499652862548828,
-0.0892333984375,
0.002750396728515625,
-0.01152801513671875,
-0.050445556640625,
0.005016326904296875,
-0.01102447509765625,
-0.04412841796875,
0.037139892578125,
0.0033016204833984375,
-0.04541015625,
0.0205535888671875,
-0.05645751953125,
0.050384521484375,
-0.003208160400390625,
-0.0323486328125,
-0.026092529296875,
-0.032806396484375,
0.038665771484375,
0.0114898681640625,
0.0237274169921875,
-0.0127105712890625,
-0.00846099853515625,
0.034149169921875,
-0.0236358642578125,
0.052032470703125,
-0.033233642578125,
-0.007427215576171875,
0.0261077880859375,
0.01187896728515625,
0.0345458984375,
0.01372528076171875,
0.03753662109375,
0.0003077983856201172,
-0.00594329833984375,
-0.0277252197265625,
-0.037109375,
0.05084228515625,
-0.07061767578125,
-0.023956298828125,
-0.0275421142578125,
-0.030975341796875,
0.0138397216796875,
-0.01959228515625,
0.036712646484375,
0.004486083984375,
-0.0185699462890625,
-0.013092041015625,
0.051513671875,
-0.018585205078125,
0.027099609375,
0.0225982666015625,
-0.0313720703125,
-0.030487060546875,
0.04632568359375,
0.0010662078857421875,
0.03558349609375,
-0.004650115966796875,
0.007656097412109375,
-0.037445068359375,
-0.049530029296875,
-0.04730224609375,
0.00833892822265625,
-0.0428466796875,
0.00274658203125,
-0.02825927734375,
0.005184173583984375,
-0.0073089599609375,
-0.00925445556640625,
-0.01309967041015625,
-0.013763427734375,
-0.053802490234375,
-0.0048675537109375,
0.05340576171875,
0.060302734375,
-0.0092926025390625,
0.0426025390625,
-0.00705718994140625,
0.03173828125,
0.0214996337890625,
0.0116729736328125,
-0.00449371337890625,
-0.0498046875,
0.0003712177276611328,
0.0023403167724609375,
-0.048797607421875,
-0.0635986328125,
0.0457763671875,
-0.004535675048828125,
0.025665283203125,
0.036529541015625,
-0.0364990234375,
0.08734130859375,
-0.0316162109375,
0.06365966796875,
0.0173797607421875,
-0.058837890625,
0.0237274169921875,
-0.05810546875,
0.00438690185546875,
0.044189453125,
0.022491455078125,
-0.019622802734375,
-0.04791259765625,
-0.062103271484375,
-0.06646728515625,
0.0258331298828125,
0.02777099609375,
0.027099609375,
0.015899658203125,
0.03564453125,
0.00033593177795410156,
0.00824737548828125,
-0.0858154296875,
-0.03997802734375,
-0.0264892578125,
0.027862548828125,
0.002498626708984375,
0.01177978515625,
-0.021575927734375,
-0.03436279296875,
0.059600830078125,
0.0244903564453125,
0.036163330078125,
0.01190948486328125,
0.0141143798828125,
-0.01442718505859375,
0.0007524490356445312,
0.06158447265625,
0.067626953125,
-0.03387451171875,
-0.0167388916015625,
-0.019134521484375,
-0.047760009765625,
0.0193939208984375,
-0.01861572265625,
-0.04791259765625,
0.01027679443359375,
-0.03509521484375,
0.05328369140625,
0.0012712478637695312,
-0.034637451171875,
0.0278472900390625,
0.005584716796875,
-0.015960693359375,
-0.027008056640625,
0.0007052421569824219,
0.010986328125,
0.03466796875,
0.031768798828125,
0.0279693603515625,
0.0276641845703125,
-0.04541015625,
-0.0037841796875,
0.045684814453125,
-0.0091400146484375,
-0.037139892578125,
0.08538818359375,
0.00518035888671875,
-0.00984954833984375,
0.01233673095703125,
-0.019073486328125,
-0.00675201416015625,
0.05950927734375,
0.043670654296875,
0.07281494140625,
0.010498046875,
0.0478515625,
0.0310516357421875,
0.010467529296875,
-0.0237884521484375,
0.0278472900390625,
0.030303955078125,
-0.0271148681640625,
-0.035980224609375,
-0.003444671630859375,
-0.0228424072265625,
0.033477783203125,
-0.0248260498046875,
0.027557373046875,
-0.06549072265625,
-0.02911376953125,
-0.0091400146484375,
-0.01166534423828125,
-0.027740478515625,
0.02325439453125,
0.0037593841552734375,
0.057037353515625,
-0.08099365234375,
0.0355224609375,
0.046661376953125,
-0.062744140625,
-0.055755615234375,
-0.00588226318359375,
0.0003094673156738281,
-0.0364990234375,
0.050445556640625,
-0.012298583984375,
-0.029327392578125,
0.006134033203125,
-0.066650390625,
-0.061065673828125,
0.09771728515625,
0.0386962890625,
-0.016693115234375,
-0.0222930908203125,
-0.017791748046875,
0.048309326171875,
-0.043365478515625,
0.037567138671875,
0.01800537109375,
0.04437255859375,
0.02362060546875,
-0.0350341796875,
0.015716552734375,
0.0033702850341796875,
0.0166015625,
-0.00536346435546875,
-0.056732177734375,
0.059722900390625,
-0.0082855224609375,
-0.04461669921875,
0.033538818359375,
0.046539306640625,
0.03460693359375,
0.018096923828125,
0.041351318359375,
0.0677490234375,
0.02215576171875,
-0.031951904296875,
0.0687255859375,
-0.002483367919921875,
0.04608154296875,
0.02874755859375,
0.00856781005859375,
0.0300140380859375,
0.02386474609375,
-0.0180206298828125,
0.06903076171875,
0.06195068359375,
-0.01107025146484375,
0.0760498046875,
0.017120361328125,
-0.0198516845703125,
-0.01235198974609375,
0.00925445556640625,
-0.040557861328125,
0.0012979507446289062,
-0.0013990402221679688,
-0.0309295654296875,
-0.03216552734375,
0.00383758544921875,
-0.001392364501953125,
-0.00710296630859375,
-0.016510009765625,
0.036407470703125,
-0.02392578125,
-0.01519012451171875,
0.033905029296875,
0.0034275054931640625,
0.05426025390625,
-0.045379638671875,
-0.003856658935546875,
-0.0208740234375,
-0.01020050048828125,
-0.0289459228515625,
-0.061431884765625,
0.0197601318359375,
0.0006394386291503906,
-0.007568359375,
-0.037994384765625,
0.0184478759765625,
-0.0006422996520996094,
-0.017669677734375,
0.020263671875,
0.0311126708984375,
0.0208587646484375,
0.0025234222412109375,
-0.08294677734375,
-0.005184173583984375,
-0.004383087158203125,
-0.007251739501953125,
0.018096923828125,
0.0132904052734375,
0.0139007568359375,
0.051727294921875,
0.038055419921875,
0.01535797119140625,
0.006069183349609375,
-0.00667572021484375,
0.060546875,
-0.048583984375,
-0.04071044921875,
-0.060577392578125,
0.051788330078125,
-0.00004088878631591797,
-0.037384033203125,
0.0592041015625,
0.06304931640625,
0.043548583984375,
-0.053466796875,
0.045257568359375,
-0.039337158203125,
0.006977081298828125,
-0.038604736328125,
0.060211181640625,
-0.05596923828125,
0.0031833648681640625,
-0.04351806640625,
-0.06060791015625,
0.0032024383544921875,
0.055938720703125,
-0.0207366943359375,
0.01367950439453125,
0.01934814453125,
0.059967041015625,
-0.0143585205078125,
-0.0204620361328125,
0.0028514862060546875,
-0.01190948486328125,
0.025421142578125,
0.054107666015625,
0.07672119140625,
-0.0284271240234375,
0.0372314453125,
-0.020416259765625,
-0.01280975341796875,
0.00400543212890625,
-0.06915283203125,
-0.0611572265625,
-0.039947509765625,
-0.0379638671875,
-0.043060302734375,
0.011260986328125,
0.0273590087890625,
0.0765380859375,
-0.040191650390625,
-0.0240020751953125,
-0.029205322265625,
0.00045871734619140625,
-0.0215606689453125,
-0.0199127197265625,
0.00324249267578125,
0.0426025390625,
-0.0767822265625,
0.01244354248046875,
0.02764892578125,
0.038421630859375,
-0.03582763671875,
0.01219940185546875,
-0.018157958984375,
-0.0004935264587402344,
0.0240020751953125,
0.03363037109375,
-0.06561279296875,
-0.034698486328125,
0.001125335693359375,
-0.01239776611328125,
0.047027587890625,
0.0604248046875,
-0.051300048828125,
0.00852203369140625,
0.040496826171875,
0.01204681396484375,
0.06451416015625,
0.029144287109375,
0.031463623046875,
-0.018157958984375,
-0.00311279296875,
0.0009570121765136719,
0.004108428955078125,
0.00829315185546875,
-0.038787841796875,
0.033935546875,
0.0275421142578125,
-0.048858642578125,
-0.02716064453125,
0.0171966552734375,
-0.090087890625,
-0.01427459716796875,
0.07061767578125,
-0.005405426025390625,
-0.003658294677734375,
-0.01293182373046875,
-0.03717041015625,
-0.004543304443359375,
-0.045257568359375,
0.0814208984375,
0.0675048828125,
-0.01885986328125,
-0.0219268798828125,
-0.073486328125,
0.0489501953125,
-0.0012226104736328125,
-0.08612060546875,
-0.007488250732421875,
0.072998046875,
0.049652099609375,
0.042144775390625,
0.043304443359375,
-0.02447509765625,
0.0227813720703125,
-0.02606201171875,
0.0189666748046875,
0.00701904296875,
-0.004058837890625,
-0.022705078125,
0.01485443115234375,
-0.01861572265625,
-0.0169219970703125
]
] |
Lykon/dreamshaper-7 | 2023-08-26T16:49:11.000Z | [
"diffusers",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"art",
"artistic",
"anime",
"dreamshaper",
"en",
"license:creativeml-openrail-m",
"endpoints_compatible",
"diffusers:StableDiffusionPipeline",
"region:us",
"has_space"
] | text-to-image | Lykon | null | null | Lykon/dreamshaper-7 | 12 | 7,519 | diffusers | 2023-08-26T16:49:11 | ---
language:
- en
license: creativeml-openrail-m
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
- art
- artistic
- diffusers
- anime
- dreamshaper
duplicated_from: lykon-models/dreamshaper-7
---
# Dreamshaper 7
`lykon-models/dreamshaper-7` is a Stable Diffusion model that has been fine-tuned on [runwayml/stable-diffusion-v1-5](https://huggingface.co/runwayml/stable-diffusion-v1-5).
Please consider supporting me:
- on [Patreon](https://www.patreon.com/Lykon275)
- or [buy me a coffee](https://snipfeed.co/lykon)
## Diffusers
For more general information on how to run text-to-image models with 🧨 Diffusers, see [the docs](https://huggingface.co/docs/diffusers/using-diffusers/conditional_image_generation).
1. Installation
```
pip install diffusers transformers accelerate
```
2. Run
```py
from diffusers import AutoPipelineForText2Image, DEISMultistepScheduler
import torch
pipe = AutoPipelineForText2Image.from_pretrained('lykon-models/dreamshaper-7', torch_dtype=torch.float16, variant="fp16")
pipe.scheduler = DEISMultistepScheduler.from_config(pipe.scheduler.config)
pipe = pipe.to("cuda")
prompt = "portrait photo of muscular bearded guy in a worn mech suit, light bokeh, intricate, steel metal, elegant, sharp focus, soft lighting, vibrant colors"
generator = torch.manual_seed(33)
image = pipe(prompt, generator=generator, num_inference_steps=25).images[0]
image.save("./image.png")
```
## Notes
- **Version 8** focuses on improving what V7 started. Might be harder to do photorealism compared to realism focused models, as it might be hard to do anime compared to anime focused models, but it can do both pretty well if you're skilled enough. Check the examples!
- **Version 7** improves lora support, NSFW and realism. If you're interested in "absolute" realism, try AbsoluteReality.
- **Version 6** adds more lora support and more style in general. It should also be better at generating directly at 1024 height (but be careful with it). 6.x are all improvements.
- **Version 5** is the best at photorealism and has noise offset.
- **Version 4** is much better with anime (can do them with no LoRA) and booru tags. It might be harder to control if you're used to caption style, so you might still want to use version 3.31. V4 is also better with eyes at lower resolutions. Overall is like a "fix" of V3 and shouldn't be too much different.
| 2,412 | [
[
-0.0261383056640625,
-0.041748046875,
0.041473388671875,
0.0214691162109375,
-0.041046142578125,
-0.01114654541015625,
0.007770538330078125,
-0.0499267578125,
0.02783203125,
0.0467529296875,
-0.0318603515625,
-0.032379150390625,
-0.05126953125,
0.006252288818359375,
-0.02984619140625,
0.051361083984375,
-0.007396697998046875,
-0.01471710205078125,
0.003570556640625,
0.01181793212890625,
-0.043914794921875,
-0.0020046234130859375,
-0.06561279296875,
-0.027099609375,
0.06536865234375,
-0.00041794776916503906,
0.057220458984375,
0.050537109375,
0.046356201171875,
0.0301666259765625,
-0.00867462158203125,
-0.005146026611328125,
-0.038238525390625,
0.01149749755859375,
0.00445556640625,
-0.028717041015625,
-0.041351318359375,
-0.023712158203125,
0.040283203125,
0.00775909423828125,
-0.0166015625,
0.0009503364562988281,
-0.00616455078125,
0.05035400390625,
-0.038238525390625,
0.009613037109375,
-0.022918701171875,
0.0082550048828125,
-0.016754150390625,
0.01812744140625,
-0.00994110107421875,
-0.01224517822265625,
-0.01012420654296875,
-0.0665283203125,
0.0203704833984375,
-0.0011005401611328125,
0.07769775390625,
0.036529541015625,
-0.0226287841796875,
0.01079559326171875,
-0.052398681640625,
0.05181884765625,
-0.0726318359375,
0.026641845703125,
0.00472259521484375,
0.042572021484375,
-0.01200103759765625,
-0.09375,
-0.036407470703125,
0.009063720703125,
-0.00212860107421875,
0.0298614501953125,
-0.015411376953125,
0.0150299072265625,
0.041961669921875,
0.03778076171875,
-0.03558349609375,
-0.004863739013671875,
-0.067626953125,
-0.012847900390625,
0.046112060546875,
-0.0014200210571289062,
0.0234832763671875,
-0.0218658447265625,
-0.01611328125,
-0.00676727294921875,
-0.0265045166015625,
-0.00717926025390625,
0.034332275390625,
-0.0084075927734375,
-0.033233642578125,
0.051177978515625,
-0.01495361328125,
0.040802001953125,
0.013763427734375,
-0.0270538330078125,
0.021881103515625,
0.02386474609375,
-0.01922607421875,
-0.01013946533203125,
0.059906005859375,
0.053009033203125,
-0.0021953582763671875,
0.0171966552734375,
-0.020233154296875,
0.0343017578125,
0.01552581787109375,
-0.081298828125,
-0.031829833984375,
0.03375244140625,
-0.04522705078125,
-0.048004150390625,
-0.025390625,
-0.04522705078125,
-0.025909423828125,
-0.00592041015625,
0.032684326171875,
-0.03631591796875,
-0.040191650390625,
0.0205078125,
-0.01953125,
0.017974853515625,
0.052032470703125,
-0.041748046875,
0.0074615478515625,
0.0269622802734375,
0.07763671875,
-0.00673675537109375,
-0.0037364959716796875,
-0.00939178466796875,
-0.003810882568359375,
-0.041046142578125,
0.070556640625,
-0.01332855224609375,
-0.029266357421875,
0.0032291412353515625,
0.00714111328125,
0.0133819580078125,
-0.039581298828125,
0.054229736328125,
-0.035888671875,
0.0198516845703125,
-0.00811004638671875,
-0.038238525390625,
-0.019622802734375,
0.0045166015625,
-0.0518798828125,
0.0770263671875,
0.0197906494140625,
-0.06640625,
0.019500732421875,
-0.0567626953125,
-0.0034942626953125,
0.0069122314453125,
-0.005519866943359375,
-0.043212890625,
0.01415252685546875,
-0.006549835205078125,
0.029541015625,
0.0033321380615234375,
0.005733489990234375,
-0.033477783203125,
-0.029998779296875,
-0.0122222900390625,
-0.0174407958984375,
0.06341552734375,
0.0280303955078125,
-0.021728515625,
0.01416778564453125,
-0.0509033203125,
-0.002269744873046875,
0.022735595703125,
-0.0023059844970703125,
0.00701141357421875,
-0.005794525146484375,
0.0211029052734375,
0.0197906494140625,
0.01360321044921875,
-0.0438232421875,
0.0164794921875,
-0.032958984375,
0.0183563232421875,
0.058990478515625,
0.00397491455078125,
0.01177978515625,
-0.04888916015625,
0.05218505859375,
0.01467132568359375,
0.018890380859375,
0.003658294677734375,
-0.058319091796875,
-0.08416748046875,
-0.027069091796875,
0.0009112358093261719,
0.0164794921875,
-0.06549072265625,
0.0189666748046875,
-0.00701904296875,
-0.0543212890625,
-0.0248260498046875,
-0.00931549072265625,
0.0250091552734375,
0.03350830078125,
0.01233673095703125,
-0.0185394287109375,
-0.041778564453125,
-0.06829833984375,
-0.001598358154296875,
0.0056304931640625,
0.0112762451171875,
-0.0089874267578125,
0.038055419921875,
-0.0174102783203125,
0.0386962890625,
-0.044189453125,
-0.0264892578125,
-0.032440185546875,
-0.0240325927734375,
0.046875,
0.0445556640625,
0.0660400390625,
-0.05596923828125,
-0.055023193359375,
-0.0010232925415039062,
-0.056793212890625,
0.0099029541015625,
0.0194549560546875,
-0.021240234375,
0.00997161865234375,
0.01146697998046875,
-0.068603515625,
0.0484619140625,
0.04937744140625,
-0.034881591796875,
0.046966552734375,
-0.01611328125,
0.01316070556640625,
-0.09857177734375,
0.007488250732421875,
0.01520538330078125,
-0.045379638671875,
-0.039703369140625,
0.013824462890625,
-0.00875091552734375,
-0.0164337158203125,
-0.059814453125,
0.058074951171875,
-0.02789306640625,
0.01065826416015625,
-0.024200439453125,
-0.00951385498046875,
0.0028667449951171875,
0.031280517578125,
0.01446533203125,
0.0248260498046875,
0.0670166015625,
-0.0372314453125,
0.04595947265625,
0.0225677490234375,
-0.039093017578125,
0.038238525390625,
-0.06756591796875,
0.02545166015625,
-0.025299072265625,
0.023193359375,
-0.056304931640625,
-0.038116455078125,
0.0491943359375,
-0.03387451171875,
0.0088958740234375,
-0.030731201171875,
-0.03057861328125,
-0.044708251953125,
-0.0238189697265625,
0.0211639404296875,
0.0838623046875,
-0.0280609130859375,
0.053375244140625,
0.0006575584411621094,
0.0190887451171875,
-0.041107177734375,
-0.038177490234375,
-0.01058197021484375,
-0.038238525390625,
-0.04931640625,
0.028076171875,
-0.0233001708984375,
-0.0204620361328125,
0.00409698486328125,
-0.005619049072265625,
0.003871917724609375,
-0.0271453857421875,
0.024078369140625,
0.0180816650390625,
-0.0175323486328125,
-0.040679931640625,
0.01025390625,
-0.01544189453125,
-0.00044846534729003906,
-0.006195068359375,
0.037750244140625,
-0.005794525146484375,
-0.00885009765625,
-0.04974365234375,
0.0085601806640625,
0.03814697265625,
0.01690673828125,
0.045684814453125,
0.057952880859375,
-0.05078125,
-0.01302337646484375,
-0.0513916015625,
-0.01056671142578125,
-0.04437255859375,
0.0226593017578125,
-0.040252685546875,
-0.0325927734375,
0.0438232421875,
0.015655517578125,
0.0298919677734375,
0.04852294921875,
0.0321044921875,
-0.032501220703125,
0.08258056640625,
0.04925537109375,
0.01776123046875,
0.03778076171875,
-0.061248779296875,
-0.0009641647338867188,
-0.06060791015625,
-0.0167999267578125,
-0.012451171875,
-0.04248046875,
-0.0236663818359375,
-0.037750244140625,
0.031097412109375,
0.01397705078125,
-0.01453399658203125,
0.0243072509765625,
-0.035736083984375,
0.038299560546875,
0.005107879638671875,
0.014617919921875,
0.0199127197265625,
0.0178680419921875,
0.0166015625,
-0.013824462890625,
-0.04754638671875,
-0.0293426513671875,
0.05029296875,
0.022979736328125,
0.0645751953125,
0.01355743408203125,
0.058990478515625,
0.017242431640625,
0.028289794921875,
-0.057220458984375,
0.04730224609375,
-0.01558685302734375,
-0.0521240234375,
0.005260467529296875,
-0.012847900390625,
-0.06878662109375,
0.0310821533203125,
-0.0215301513671875,
-0.046478271484375,
0.0138397216796875,
0.04632568359375,
-0.01457977294921875,
0.03277587890625,
-0.04034423828125,
0.0706787109375,
-0.004642486572265625,
-0.037078857421875,
-0.01342010498046875,
-0.033477783203125,
0.038360595703125,
0.01209259033203125,
0.01398468017578125,
-0.0303192138671875,
-0.009613037109375,
0.041015625,
-0.03533935546875,
0.06951904296875,
-0.0253143310546875,
-0.01141357421875,
0.030059814453125,
0.00835418701171875,
0.031402587890625,
0.0007543563842773438,
-0.011993408203125,
0.004100799560546875,
0.0161590576171875,
-0.0360107421875,
-0.04718017578125,
0.069091796875,
-0.061248779296875,
-0.036468505859375,
-0.04449462890625,
-0.0225677490234375,
0.01076507568359375,
0.037689208984375,
0.062286376953125,
0.0472412109375,
0.01593017578125,
0.0186309814453125,
0.044647216796875,
0.004711151123046875,
0.040924072265625,
-0.00766754150390625,
-0.063720703125,
-0.04791259765625,
0.050537109375,
0.0033016204833984375,
0.030914306640625,
0.004062652587890625,
0.034393310546875,
-0.025665283203125,
-0.0377197265625,
-0.04254150390625,
0.0318603515625,
-0.0618896484375,
-0.031951904296875,
-0.03521728515625,
-0.031768798828125,
-0.0215911865234375,
-0.0237274169921875,
-0.039703369140625,
-0.0298919677734375,
-0.02587890625,
0.0150604248046875,
0.055908203125,
0.059173583984375,
0.004978179931640625,
0.0210418701171875,
-0.039154052734375,
0.04144287109375,
0.004978179931640625,
0.0307159423828125,
-0.01042938232421875,
-0.05340576171875,
-0.010955810546875,
-0.002735137939453125,
-0.036865234375,
-0.05438232421875,
0.0386962890625,
0.00344085693359375,
0.0263671875,
0.0294952392578125,
-0.003490447998046875,
0.061370849609375,
-0.0279541015625,
0.06146240234375,
0.038330078125,
-0.04229736328125,
0.038726806640625,
-0.061065673828125,
0.019439697265625,
0.011627197265625,
0.017669677734375,
-0.03643798828125,
-0.0260009765625,
-0.07281494140625,
-0.04754638671875,
0.03173828125,
0.045318603515625,
0.0168304443359375,
0.01922607421875,
0.049285888671875,
-0.004077911376953125,
0.0157012939453125,
-0.08038330078125,
-0.028533935546875,
-0.042449951171875,
-0.0167388916015625,
0.0025806427001953125,
-0.0015573501586914062,
0.01139068603515625,
-0.02728271484375,
0.0645751953125,
-0.012908935546875,
0.0299530029296875,
0.032470703125,
0.0223541259765625,
-0.0213775634765625,
-0.0157012939453125,
0.0298919677734375,
0.0518798828125,
-0.023468017578125,
0.00399017333984375,
0.001438140869140625,
-0.034210205078125,
0.00727081298828125,
-0.005634307861328125,
-0.0290374755859375,
0.0191650390625,
0.0130462646484375,
0.0794677734375,
0.0018377304077148438,
-0.028289794921875,
0.032318115234375,
0.0014896392822265625,
-0.02374267578125,
-0.038818359375,
0.032318115234375,
0.007358551025390625,
0.0248260498046875,
-0.00634765625,
0.041046142578125,
0.020355224609375,
-0.0300445556640625,
-0.0239105224609375,
0.0133209228515625,
-0.027587890625,
-0.037139892578125,
0.07073974609375,
0.01105499267578125,
-0.033477783203125,
0.014862060546875,
-0.0138702392578125,
-0.0180206298828125,
0.05224609375,
0.0625,
0.060302734375,
-0.005397796630859375,
0.014923095703125,
0.0498046875,
0.001926422119140625,
-0.033294677734375,
0.048065185546875,
0.0278472900390625,
-0.04248046875,
-0.0206756591796875,
-0.0377197265625,
-0.017669677734375,
0.02587890625,
-0.02923583984375,
0.0609130859375,
-0.041839599609375,
-0.0291748046875,
-0.01236724853515625,
-0.00624847412109375,
-0.05255126953125,
0.0265045166015625,
0.0156402587890625,
0.06927490234375,
-0.04248046875,
0.06341552734375,
0.05682373046875,
-0.037322998046875,
-0.057708740234375,
-0.0182647705078125,
-0.00666046142578125,
-0.03515625,
0.005584716796875,
0.0173492431640625,
-0.01274871826171875,
0.0205841064453125,
-0.04656982421875,
-0.06866455078125,
0.09027099609375,
0.03778076171875,
-0.048095703125,
-0.01215362548828125,
-0.01885986328125,
0.03924560546875,
-0.0229339599609375,
0.00211334228515625,
0.01378631591796875,
0.0245361328125,
0.0200347900390625,
-0.053192138671875,
-0.01166534423828125,
-0.032379150390625,
0.03057861328125,
0.005367279052734375,
-0.06842041015625,
0.04693603515625,
-0.0300445556640625,
-0.00921630859375,
0.0523681640625,
0.0699462890625,
0.038116455078125,
0.0133514404296875,
0.04638671875,
0.054290771484375,
0.0196685791015625,
-0.004016876220703125,
0.0836181640625,
-0.004917144775390625,
0.031707763671875,
0.053009033203125,
-0.0023860931396484375,
0.06365966796875,
0.033203125,
-0.01319122314453125,
0.044830322265625,
0.054412841796875,
-0.003856658935546875,
0.03692626953125,
0.005809783935546875,
-0.016265869140625,
-0.0120849609375,
0.002529144287109375,
-0.043212890625,
-0.00222015380859375,
0.01102447509765625,
-0.012908935546875,
0.003314971923828125,
0.006465911865234375,
0.004299163818359375,
-0.009674072265625,
-0.0246429443359375,
0.03155517578125,
0.023895263671875,
-0.033966064453125,
0.04949951171875,
-0.009765625,
0.06640625,
-0.059906005859375,
-0.0121307373046875,
-0.01953125,
0.0302581787109375,
-0.03387451171875,
-0.06610107421875,
0.0144805908203125,
-0.003292083740234375,
0.00298309326171875,
-0.0182952880859375,
0.07684326171875,
-0.0158538818359375,
-0.06622314453125,
0.0178070068359375,
0.0006594657897949219,
0.0260009765625,
0.0036296844482421875,
-0.059906005859375,
0.0430908203125,
0.00420379638671875,
-0.02197265625,
0.0004286766052246094,
0.0036411285400390625,
0.020416259765625,
0.050811767578125,
0.031494140625,
0.0161590576171875,
0.012481689453125,
0.0132904052734375,
0.054046630859375,
-0.0296630859375,
-0.03155517578125,
-0.040313720703125,
0.065673828125,
-0.016448974609375,
-0.0262298583984375,
0.0496826171875,
0.055908203125,
0.047515869140625,
-0.043243408203125,
0.0596923828125,
-0.02447509765625,
0.029327392578125,
-0.047637939453125,
0.049560546875,
-0.058441162109375,
0.00691986083984375,
-0.03466796875,
-0.09393310546875,
-0.018310546875,
0.07135009765625,
0.004486083984375,
0.014434814453125,
0.040557861328125,
0.058807373046875,
-0.030914306640625,
-0.0250091552734375,
0.0372314453125,
0.0155181884765625,
0.028472900390625,
0.0151519775390625,
0.07891845703125,
-0.0675048828125,
0.0280303955078125,
-0.06195068359375,
0.0043182373046875,
0.008026123046875,
-0.0704345703125,
-0.06805419921875,
-0.04156494140625,
-0.052978515625,
-0.051177978515625,
0.00453948974609375,
0.054779052734375,
0.07122802734375,
-0.0278778076171875,
-0.0200958251953125,
-0.00699615478515625,
-0.00675201416015625,
-0.001972198486328125,
-0.015045166015625,
0.0104827880859375,
0.006622314453125,
-0.0740966796875,
0.01259613037109375,
0.01213836669921875,
0.0318603515625,
-0.0216064453125,
-0.0140228271484375,
0.0125274658203125,
-0.00995635986328125,
0.03851318359375,
0.0140380859375,
-0.06646728515625,
-0.02569580078125,
-0.01042938232421875,
0.018157958984375,
0.0194244384765625,
0.037139892578125,
-0.0498046875,
0.0157012939453125,
0.033203125,
-0.0010280609130859375,
0.05462646484375,
-0.020721435546875,
0.0225067138671875,
-0.024566650390625,
0.026885986328125,
0.0146636962890625,
0.0478515625,
0.02374267578125,
-0.0244140625,
0.037445068359375,
0.039154052734375,
-0.043304443359375,
-0.05645751953125,
0.0144500732421875,
-0.09783935546875,
-0.0008020401000976562,
0.084716796875,
0.0166015625,
-0.02545166015625,
0.00970458984375,
-0.05657958984375,
-0.0032901763916015625,
-0.020416259765625,
0.041351318359375,
0.0287017822265625,
-0.01425933837890625,
-0.03631591796875,
-0.050537109375,
0.04327392578125,
-0.0023822784423828125,
-0.0643310546875,
-0.021240234375,
0.0478515625,
0.048980712890625,
0.0183563232421875,
0.06695556640625,
-0.01161956787109375,
0.0210418701171875,
0.003997802734375,
0.004802703857421875,
0.01190948486328125,
-0.0166015625,
-0.01125335693359375,
-0.01140594482421875,
0.00972747802734375,
-0.00821685791015625
]
] |
PygmalionAI/pygmalion-1.3b | 2023-06-27T05:36:50.000Z | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"gpt_neox",
"text-generation",
"text generation",
"conversational",
"en",
"license:agpl-3.0",
"has_space",
"text-generation-inference",
"region:us"
] | conversational | PygmalionAI | null | null | PygmalionAI/pygmalion-1.3b | 54 | 7,512 | transformers | 2022-12-25T17:54:27 | ---
license: agpl-3.0
language:
- en
thumbnail:
tags:
- text generation
- conversational
inference: false
---
# Pygmalion 1.3B
## Model description
Pymalion 1.3B is a proof-of-concept dialogue model based on EleutherAI's [pythia-1.3b-deduped](https://huggingface.co/EleutherAI/pythia-1.3b-deduped).
**Warning:** This model is **NOT** suitable for use by minors. It **will** output X-rated content under certain circumstances.
## Training data
The fine-tuning dataset consisted of 56MB of dialogue data gathered from multiple sources, which includes both real _and_ partially machine-generated conversations.
## Training procedure
Fine-tuning was done using [ColossalAI](https://github.com/hpcaitech/ColossalAI) (specifically, with a slightly modified version of their [OPT fine-tune example](https://github.com/hpcaitech/ColossalAI/blob/78509124d32b63b7fc36f6508e0576a326d51422/examples/language/opt/run_clm.py)) for around 11.4 million tokens over 5440 steps on a single 24GB GPU. The run took just under 21 hours.
## Intended use
### The easy way
We provide a notebook with a Gradio UI for playing around with the model without having to manually format inputs. This notebook can be found [here](https://github.com/PygmalionAI/gradio-ui/blob/master/notebooks/GPU.ipynb).
### The manual way
The model can be used as a regular text generation model, but it'll perform best if the input prompt adheres to the following format:
```
[CHARACTER]'s Persona: [A few sentences about the character you want the model to play]
[DIALOGUE HISTORY]
You: [Your input message here]
[CHARACTER]:
```
Where `[CHARACTER] `is, as you can probably guess, the name of the character you want the model to portray, and `[DIALOGUE HISTORY]` is chat history so the model can have some conversational context to draw from. Ideally it'll be pairs of messages like:
```
[CHARACTER]: [some dialogue here]
You: [your response to the dialogue above]
```
Apart from chat history, you can also just add example conversations in `[DIALOGUE HISTORY]` to show how the character should speak - ideally at the beginning, so it doesn't get confused as to what's conversation history vs. character definition.
## Known issues
- The model can get stuck repeating certain phrases, or sometimes even entire sentences.
- We believe this is due to that behavior being present in the training data itself, and plan to investigate and adjust accordingly for future versions.
| 2,452 | [
[
-0.0216064453125,
-0.0704345703125,
0.0203704833984375,
0.009002685546875,
-0.034423828125,
-0.024871826171875,
-0.006046295166015625,
-0.0288238525390625,
0.01508331298828125,
0.034088134765625,
-0.06683349609375,
-0.0227508544921875,
-0.0264739990234375,
-0.002033233642578125,
0.0009670257568359375,
0.0972900390625,
0.01496124267578125,
-0.0020389556884765625,
-0.0014390945434570312,
0.01343536376953125,
-0.03759765625,
-0.0308074951171875,
-0.06561279296875,
-0.040496826171875,
0.047576904296875,
0.025238037109375,
0.05743408203125,
0.041351318359375,
0.0012388229370117188,
0.0207672119140625,
-0.0275421142578125,
0.002613067626953125,
-0.0523681640625,
-0.00238037109375,
-0.00848388671875,
-0.030731201171875,
-0.036224365234375,
0.008880615234375,
0.050872802734375,
0.02490234375,
-0.00598907470703125,
0.01316070556640625,
-0.005474090576171875,
0.0130462646484375,
-0.0294647216796875,
0.0236663818359375,
-0.032806396484375,
-0.0025119781494140625,
-0.01003265380859375,
0.000995635986328125,
-0.01203155517578125,
-0.0213623046875,
0.0296173095703125,
-0.06158447265625,
0.00852203369140625,
0.0027370452880859375,
0.078125,
-0.004467010498046875,
-0.0238800048828125,
-0.017547607421875,
-0.035247802734375,
0.0498046875,
-0.077880859375,
-0.0151519775390625,
0.0311737060546875,
0.0178375244140625,
-0.0118865966796875,
-0.06512451171875,
-0.043701171875,
-0.0212554931640625,
-0.00858306884765625,
0.007537841796875,
-0.01522064208984375,
0.0237274169921875,
0.0491943359375,
0.042236328125,
-0.050506591796875,
-0.00614166259765625,
-0.0382080078125,
-0.007587432861328125,
0.030853271484375,
0.0309600830078125,
0.031829833984375,
-0.0293731689453125,
-0.004627227783203125,
-0.00717926025390625,
-0.036376953125,
0.0115966796875,
0.045928955078125,
0.00978851318359375,
-0.01279449462890625,
0.036651611328125,
-0.01146697998046875,
0.045318603515625,
0.0190582275390625,
-0.0291900634765625,
0.004192352294921875,
-0.01483917236328125,
-0.0248870849609375,
0.0158233642578125,
0.080322265625,
0.036590576171875,
0.008880615234375,
0.0154266357421875,
0.005527496337890625,
-0.0001800060272216797,
0.0172882080078125,
-0.09149169921875,
-0.0440673828125,
0.023956298828125,
-0.03363037109375,
-0.0350341796875,
-0.024810791015625,
-0.0430908203125,
-0.0272674560546875,
-0.00408172607421875,
0.0270233154296875,
-0.051849365234375,
-0.025909423828125,
-0.007465362548828125,
-0.00795745849609375,
-0.00505828857421875,
0.027587890625,
-0.085205078125,
0.01303863525390625,
0.03753662109375,
0.071044921875,
0.017730712890625,
-0.0313720703125,
-0.0185699462890625,
-0.00797271728515625,
-0.0164947509765625,
0.0310516357421875,
-0.01824951171875,
-0.0311431884765625,
-0.015655517578125,
0.0105743408203125,
-0.0206298828125,
-0.0222930908203125,
0.04327392578125,
-0.0024662017822265625,
0.044158935546875,
0.0127105712890625,
-0.045684814453125,
-0.033538818359375,
0.0098419189453125,
-0.045257568359375,
0.04766845703125,
0.02264404296875,
-0.07330322265625,
0.004291534423828125,
-0.039031982421875,
-0.021270751953125,
0.016632080078125,
-0.0048370361328125,
-0.0279388427734375,
0.005218505859375,
0.01441192626953125,
0.0223541259765625,
-0.034912109375,
0.03369140625,
-0.015472412109375,
-0.04071044921875,
0.027618408203125,
-0.029327392578125,
0.06854248046875,
0.022674560546875,
-0.0227813720703125,
-0.005199432373046875,
-0.0390625,
0.004451751708984375,
0.01299285888671875,
-0.0180511474609375,
0.003513336181640625,
-0.002826690673828125,
-0.0021209716796875,
0.004886627197265625,
0.026153564453125,
-0.0341796875,
0.0202484130859375,
-0.042144775390625,
0.039886474609375,
0.03546142578125,
0.009735107421875,
0.024200439453125,
-0.053070068359375,
0.043212890625,
-0.007282257080078125,
0.0169525146484375,
-0.033905029296875,
-0.05615234375,
-0.056182861328125,
-0.027740478515625,
0.0206756591796875,
0.0489501953125,
-0.053619384765625,
0.023681640625,
0.0025768280029296875,
-0.048583984375,
-0.027252197265625,
-0.00794219970703125,
0.039703369140625,
0.044403076171875,
-0.00279998779296875,
-0.0018587112426757812,
-0.054168701171875,
-0.0673828125,
-0.017822265625,
-0.055908203125,
-0.00820159912109375,
0.038604736328125,
0.036376953125,
-0.01593017578125,
0.048248291015625,
-0.033782958984375,
0.007625579833984375,
-0.040985107421875,
0.01947021484375,
0.0360107421875,
0.05682373046875,
0.0313720703125,
-0.04400634765625,
-0.0189666748046875,
-0.00302886962890625,
-0.05938720703125,
-0.01080322265625,
-0.020538330078125,
-0.00628662109375,
-0.0031528472900390625,
0.006755828857421875,
-0.062744140625,
0.0318603515625,
0.03955078125,
-0.03216552734375,
0.037017822265625,
-0.006473541259765625,
0.018768310546875,
-0.10333251953125,
0.0187530517578125,
0.006847381591796875,
-0.017578125,
-0.055084228515625,
0.009979248046875,
-0.02117919921875,
-0.031097412109375,
-0.039794921875,
0.049072265625,
-0.00934600830078125,
0.0309906005859375,
-0.0028743743896484375,
0.00838470458984375,
-0.0062408447265625,
0.049591064453125,
0.0019969940185546875,
0.045623779296875,
0.041259765625,
-0.05535888671875,
0.045166015625,
0.03265380859375,
-0.03472900390625,
0.0278167724609375,
-0.0770263671875,
0.02203369140625,
0.0055084228515625,
0.0167083740234375,
-0.06878662109375,
-0.033935546875,
0.058074951171875,
-0.060211181640625,
0.020965576171875,
-0.04534912109375,
-0.03369140625,
-0.021942138671875,
-0.0035343170166015625,
0.0283966064453125,
0.04962158203125,
-0.0382080078125,
0.059051513671875,
0.02764892578125,
-0.027008056640625,
-0.0140228271484375,
-0.038055419921875,
0.0010023117065429688,
-0.039093017578125,
-0.062164306640625,
0.01459503173828125,
-0.01197052001953125,
-0.007366180419921875,
-0.02923583984375,
0.01092529296875,
-0.006061553955078125,
0.0130767822265625,
0.030029296875,
0.01861572265625,
0.001605987548828125,
-0.009063720703125,
0.01151275634765625,
0.0017118453979492188,
-0.0026397705078125,
-0.009307861328125,
0.04931640625,
-0.0022525787353515625,
0.01039886474609375,
-0.06402587890625,
0.01082611083984375,
0.047271728515625,
-0.0014553070068359375,
0.033447265625,
0.039764404296875,
-0.034393310546875,
0.0165252685546875,
-0.0193023681640625,
0.0003116130828857422,
-0.032379150390625,
0.035797119140625,
-0.033447265625,
-0.04876708984375,
0.0399169921875,
-0.006134033203125,
0.00539398193359375,
0.0255279541015625,
0.040283203125,
0.002231597900390625,
0.10430908203125,
0.022064208984375,
-0.0002796649932861328,
0.055694580078125,
-0.020538330078125,
0.002338409423828125,
-0.0703125,
-0.0255279541015625,
-0.0251312255859375,
-0.0160369873046875,
-0.052215576171875,
-0.0146484375,
0.014190673828125,
0.0263824462890625,
-0.0262603759765625,
0.0421142578125,
-0.022247314453125,
0.0274658203125,
0.041748046875,
0.01335906982421875,
-0.0014715194702148438,
0.00457000732421875,
0.0076904296875,
-0.006687164306640625,
-0.06683349609375,
-0.052520751953125,
0.06866455078125,
0.0477294921875,
0.060821533203125,
0.0112457275390625,
0.05767822265625,
-0.0138702392578125,
0.008514404296875,
-0.060791015625,
0.04156494140625,
0.00027751922607421875,
-0.060821533203125,
-0.0229339599609375,
-0.04901123046875,
-0.06512451171875,
0.016845703125,
-0.0133056640625,
-0.08807373046875,
-0.003612518310546875,
0.00955963134765625,
-0.0350341796875,
0.01331329345703125,
-0.0589599609375,
0.09466552734375,
-0.007671356201171875,
-0.0261077880859375,
-0.00403594970703125,
-0.042144775390625,
0.03839111328125,
0.0262603759765625,
-0.016357421875,
0.0011806488037109375,
0.0294952392578125,
0.060516357421875,
-0.036041259765625,
0.063720703125,
-0.0111236572265625,
-0.005130767822265625,
0.038299560546875,
0.023193359375,
0.02825927734375,
0.045654296875,
0.0174407958984375,
0.004634857177734375,
0.0288238525390625,
-0.01525115966796875,
-0.037139892578125,
0.06683349609375,
-0.064208984375,
-0.035125732421875,
-0.043487548828125,
-0.044891357421875,
0.0094757080078125,
0.00833892822265625,
0.034881591796875,
0.047637939453125,
-0.03472900390625,
0.006832122802734375,
0.05694580078125,
-0.0209808349609375,
0.0250244140625,
0.029327392578125,
-0.03973388671875,
-0.0562744140625,
0.06610107421875,
-0.006862640380859375,
0.007537841796875,
0.00629425048828125,
0.026458740234375,
-0.01383209228515625,
-0.012481689453125,
-0.059356689453125,
0.0191192626953125,
-0.03546142578125,
0.00411224365234375,
-0.048583984375,
-0.005706787109375,
-0.031463623046875,
0.020263671875,
-0.01337432861328125,
-0.03448486328125,
-0.04754638671875,
0.0200347900390625,
0.02154541015625,
0.037322998046875,
0.017974853515625,
0.0440673828125,
-0.053375244140625,
0.023834228515625,
0.03546142578125,
0.00870513916015625,
-0.01131439208984375,
-0.05938720703125,
-0.0200042724609375,
0.0293426513671875,
-0.034820556640625,
-0.06768798828125,
0.03887939453125,
0.0159759521484375,
0.037109375,
0.036346435546875,
-0.016845703125,
0.045989990234375,
-0.025299072265625,
0.078125,
0.0181732177734375,
-0.06475830078125,
0.05560302734375,
-0.045562744140625,
0.0341796875,
0.038421630859375,
0.019989013671875,
-0.054473876953125,
-0.0184173583984375,
-0.0633544921875,
-0.0416259765625,
0.08123779296875,
0.041046142578125,
0.013702392578125,
-0.01012420654296875,
0.030548095703125,
-0.004779815673828125,
0.01885986328125,
-0.048248291015625,
-0.015838623046875,
-0.02960205078125,
-0.0167694091796875,
-0.01177978515625,
-0.0172576904296875,
-0.01520538330078125,
-0.031494140625,
0.06463623046875,
-0.0133056640625,
0.0231781005859375,
0.01316070556640625,
-0.007114410400390625,
-0.0076751708984375,
-0.006694793701171875,
0.046844482421875,
0.04962158203125,
-0.03668212890625,
-0.021331787109375,
0.0013370513916015625,
-0.03924560546875,
-0.0136871337890625,
0.0180511474609375,
-0.0204925537109375,
0.022247314453125,
0.0157623291015625,
0.1002197265625,
0.00897979736328125,
-0.04388427734375,
0.03118896484375,
-0.028289794921875,
-0.012725830078125,
-0.0241241455078125,
0.01332855224609375,
0.00020503997802734375,
0.03173828125,
0.00408935546875,
-0.017578125,
0.01026153564453125,
-0.06207275390625,
-0.014068603515625,
0.002246856689453125,
-0.0193328857421875,
-0.0196075439453125,
0.0489501953125,
0.0128936767578125,
-0.041107177734375,
0.053070068359375,
-0.0144500732421875,
-0.04388427734375,
0.049530029296875,
0.055908203125,
0.0655517578125,
-0.005008697509765625,
0.028564453125,
0.05364990234375,
0.01154327392578125,
-0.01491546630859375,
0.0146942138671875,
0.0029430389404296875,
-0.03265380859375,
-0.0160675048828125,
-0.031768798828125,
-0.0293731689453125,
0.0285491943359375,
-0.036468505859375,
0.02374267578125,
-0.057647705078125,
-0.0185089111328125,
-0.0077972412109375,
0.01416015625,
-0.026641845703125,
0.025634765625,
0.0211029052734375,
0.055908203125,
-0.0517578125,
0.0477294921875,
0.0643310546875,
-0.045074462890625,
-0.0634765625,
-0.01092529296875,
-0.01280975341796875,
-0.03759765625,
0.031951904296875,
0.0189666748046875,
0.0194091796875,
0.00482940673828125,
-0.059844970703125,
-0.0384521484375,
0.101806640625,
0.02825927734375,
-0.043701171875,
-0.02166748046875,
-0.0139007568359375,
0.035614013671875,
-0.04107666015625,
0.044677734375,
0.04510498046875,
0.01316070556640625,
0.0101165771484375,
-0.078125,
0.0013637542724609375,
-0.034515380859375,
0.0208740234375,
-0.004154205322265625,
-0.0478515625,
0.08984375,
0.0016946792602539062,
-0.02386474609375,
0.055908203125,
0.038787841796875,
0.03192138671875,
0.021392822265625,
0.0269775390625,
0.05029296875,
0.054656982421875,
-0.0198516845703125,
0.08807373046875,
-0.044189453125,
0.0313720703125,
0.09033203125,
0.01145172119140625,
0.0208892822265625,
0.0250396728515625,
0.00794219970703125,
0.0372314453125,
0.061981201171875,
0.0032196044921875,
0.040283203125,
0.021209716796875,
-0.0232086181640625,
-0.02410888671875,
-0.0083160400390625,
-0.028900146484375,
0.02008056640625,
0.0204620361328125,
-0.03857421875,
0.01018524169921875,
-0.012603759765625,
0.0084686279296875,
-0.0240325927734375,
-0.027099609375,
0.059722900390625,
-0.003143310546875,
-0.0592041015625,
0.052032470703125,
0.004352569580078125,
0.041717529296875,
-0.05657958984375,
-0.0183258056640625,
-0.029388427734375,
0.007381439208984375,
-0.00537872314453125,
-0.033599853515625,
-0.0253143310546875,
-0.00586700439453125,
-0.005645751953125,
0.0038700103759765625,
0.047943115234375,
-0.037017822265625,
-0.0357666015625,
-0.01554107666015625,
0.022308349609375,
0.0340576171875,
-0.01210784912109375,
-0.048828125,
0.003780364990234375,
0.003448486328125,
-0.00411224365234375,
0.01385498046875,
0.037322998046875,
0.00830841064453125,
0.0494384765625,
0.03076171875,
-0.005413055419921875,
-0.0138397216796875,
0.01316070556640625,
0.0673828125,
-0.034332275390625,
-0.04034423828125,
-0.05267333984375,
0.04351806640625,
-0.00942230224609375,
-0.059417724609375,
0.05517578125,
0.045928955078125,
0.052337646484375,
-0.0099945068359375,
0.058074951171875,
-0.0192413330078125,
0.028106689453125,
-0.045745849609375,
0.053253173828125,
-0.02716064453125,
0.0093994140625,
-0.039642333984375,
-0.06573486328125,
0.01145172119140625,
0.08294677734375,
-0.0027942657470703125,
0.00377655029296875,
0.055999755859375,
0.072998046875,
0.0037364959716796875,
0.0209808349609375,
0.022491455078125,
0.011199951171875,
0.01508331298828125,
0.0687255859375,
0.09112548828125,
-0.059906005859375,
0.0377197265625,
-0.0186004638671875,
-0.0194854736328125,
-0.0122222900390625,
-0.04974365234375,
-0.09759521484375,
-0.027099609375,
-0.0290985107421875,
-0.052825927734375,
0.009552001953125,
0.07989501953125,
0.048187255859375,
-0.042755126953125,
-0.0278472900390625,
-0.01175689697265625,
0.0008358955383300781,
-0.01045989990234375,
-0.0123748779296875,
-0.0162353515625,
0.005695343017578125,
-0.070068359375,
0.0267791748046875,
-0.0182647705078125,
0.01849365234375,
-0.0153045654296875,
-0.0195465087890625,
-0.023406982421875,
-0.00013458728790283203,
0.014617919921875,
0.02813720703125,
-0.046630859375,
-0.01433563232421875,
-0.0191497802734375,
-0.0008339881896972656,
-0.005939483642578125,
0.071044921875,
-0.053680419921875,
0.037506103515625,
0.0367431640625,
0.01031494140625,
0.058135986328125,
-0.00640106201171875,
0.0504150390625,
-0.038299560546875,
0.011962890625,
0.0166473388671875,
0.0301055908203125,
0.027069091796875,
-0.0298919677734375,
0.023193359375,
0.020721435546875,
-0.04571533203125,
-0.047149658203125,
0.0229339599609375,
-0.058807373046875,
-0.014312744140625,
0.0916748046875,
-0.022308349609375,
-0.0281982421875,
0.00865936279296875,
-0.0694580078125,
0.037841796875,
-0.041229248046875,
0.047332763671875,
0.053436279296875,
0.004161834716796875,
-0.0218048095703125,
-0.02947998046875,
0.03472900390625,
0.01611328125,
-0.040924072265625,
0.0038852691650390625,
0.051971435546875,
0.02764892578125,
0.01338958740234375,
0.04400634765625,
-0.0085601806640625,
0.0391845703125,
0.01554107666015625,
0.00119781494140625,
-0.01345062255859375,
-0.0226898193359375,
-0.0374755859375,
-0.0184478759765625,
-0.0007863044738769531,
-0.006496429443359375
]
] |
facebook/deit-small-patch16-224 | 2022-07-13T11:41:40.000Z | [
"transformers",
"pytorch",
"tf",
"vit",
"image-classification",
"dataset:imagenet-1k",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | image-classification | facebook | null | null | facebook/deit-small-patch16-224 | 2 | 7,508 | transformers | 2022-03-02T23:29:05 | ---
license: apache-2.0
tags:
- image-classification
datasets:
- imagenet-1k
---
# Data-efficient Image Transformer (small-sized model)
Data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is actually a more efficiently trained Vision Transformer (ViT).
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pre-trained and fine-tuned on a large collection of images in a supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a more efficiently trained ViT model, you can plug it into ViTModel or ViTForImageClassification. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, ViTForImageClassification
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-small-patch16-224')
model = ViTForImageClassification.from_pretrained('facebook/deit-small-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
The ViT model was pretrained on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| DeiT-tiny | 72.2 | 91.1 | 5M | https://huggingface.co/facebook/deit-tiny-patch16-224 |
| **DeiT-small** | **79.9** | **95.0** | **22M** | **https://huggingface.co/facebook/deit-small-patch16-224** |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
``` | 7,285 | [
[
-0.05706787109375,
-0.032958984375,
0.00446319580078125,
0.00269317626953125,
-0.02789306640625,
-0.01849365234375,
-0.00885009765625,
-0.0384521484375,
0.0261383056640625,
0.0172576904296875,
-0.0290069580078125,
-0.02447509765625,
-0.060546875,
0.00020742416381835938,
-0.033355712890625,
0.06787109375,
0.00238800048828125,
-0.01117706298828125,
-0.007793426513671875,
-0.01154327392578125,
-0.03253173828125,
-0.034820556640625,
-0.053497314453125,
-0.0109100341796875,
0.038360595703125,
0.01654052734375,
0.050384521484375,
0.062042236328125,
0.065185546875,
0.0350341796875,
-0.00693511962890625,
0.00551605224609375,
-0.0369873046875,
-0.024749755859375,
0.005985260009765625,
-0.02349853515625,
-0.03094482421875,
0.0169677734375,
0.03985595703125,
0.03167724609375,
0.0167083740234375,
0.0218505859375,
0.019073486328125,
0.054443359375,
-0.03753662109375,
0.01282501220703125,
-0.03424072265625,
0.017303466796875,
-0.003173828125,
-0.005035400390625,
-0.020721435546875,
-0.0171661376953125,
0.0165557861328125,
-0.03704833984375,
0.0276031494140625,
-0.01071929931640625,
0.10430908203125,
0.0302581787109375,
-0.023406982421875,
0.01366424560546875,
-0.046722412109375,
0.05389404296875,
-0.03570556640625,
0.0293731689453125,
0.0241241455078125,
0.02557373046875,
0.006618499755859375,
-0.0765380859375,
-0.038360595703125,
-0.00823974609375,
-0.0245819091796875,
0.0143585205078125,
-0.0244293212890625,
0.0033397674560546875,
0.038970947265625,
0.046722412109375,
-0.034576416015625,
0.0010242462158203125,
-0.036590576171875,
-0.0164642333984375,
0.04840087890625,
-0.0108642578125,
0.00635528564453125,
-0.00778961181640625,
-0.048126220703125,
-0.02459716796875,
-0.0170440673828125,
0.0093994140625,
0.005054473876953125,
0.003658294677734375,
-0.01715087890625,
0.035400390625,
-0.00860595703125,
0.045013427734375,
0.03558349609375,
-0.0013275146484375,
0.04241943359375,
-0.0234527587890625,
-0.0293426513671875,
-0.009033203125,
0.07098388671875,
0.034393310546875,
0.0216522216796875,
0.010345458984375,
-0.018218994140625,
0.006954193115234375,
0.020233154296875,
-0.08233642578125,
-0.0218048095703125,
-0.0014247894287109375,
-0.05718994140625,
-0.035919189453125,
0.0191650390625,
-0.04632568359375,
-0.0037841796875,
-0.0277862548828125,
0.04248046875,
-0.0300445556640625,
-0.0241851806640625,
-0.011566162109375,
-0.0033931732177734375,
0.0299835205078125,
0.0246429443359375,
-0.041961669921875,
0.01483154296875,
0.019805908203125,
0.07537841796875,
-0.00732421875,
-0.01229095458984375,
-0.0060272216796875,
-0.0275726318359375,
-0.033721923828125,
0.045928955078125,
-0.0020961761474609375,
-0.0015916824340820312,
-0.015167236328125,
0.027557373046875,
-0.006214141845703125,
-0.031982421875,
0.0244903564453125,
-0.034149169921875,
-0.0025691986083984375,
-0.019805908203125,
-0.024322509765625,
-0.017730712890625,
0.0236053466796875,
-0.0615234375,
0.0885009765625,
0.0206298828125,
-0.06890869140625,
0.03076171875,
-0.03887939453125,
-0.00566864013671875,
-0.006626129150390625,
0.0008792877197265625,
-0.04522705078125,
-0.00659942626953125,
0.02459716796875,
0.046539306640625,
-0.01425933837890625,
-0.004192352294921875,
-0.019744873046875,
-0.03692626953125,
0.01971435546875,
-0.0313720703125,
0.06903076171875,
0.0248565673828125,
-0.0382080078125,
-0.00547027587890625,
-0.05816650390625,
0.0046844482421875,
0.0267181396484375,
-0.0176544189453125,
-0.006374359130859375,
-0.033782958984375,
0.00608062744140625,
0.032501220703125,
0.0181427001953125,
-0.0440673828125,
0.007358551025390625,
-0.0106964111328125,
0.039306640625,
0.0634765625,
-0.0100860595703125,
0.0281524658203125,
-0.02264404296875,
0.022308349609375,
0.0231170654296875,
0.03338623046875,
-0.0216827392578125,
-0.033599853515625,
-0.0712890625,
-0.029083251953125,
0.034515380859375,
0.0296630859375,
-0.0523681640625,
0.048126220703125,
-0.03271484375,
-0.04901123046875,
-0.0274505615234375,
0.0036716461181640625,
0.02703857421875,
0.03662109375,
0.032257080078125,
-0.039154052734375,
-0.037994384765625,
-0.08026123046875,
0.00789642333984375,
-0.0035686492919921875,
0.01220703125,
0.0146636962890625,
0.05267333984375,
-0.0181732177734375,
0.07342529296875,
-0.03216552734375,
-0.0245208740234375,
-0.0016698837280273438,
-0.00612640380859375,
0.025054931640625,
0.05267333984375,
0.06427001953125,
-0.0732421875,
-0.0517578125,
0.0027103424072265625,
-0.06219482421875,
0.017852783203125,
0.0009441375732421875,
-0.0287322998046875,
0.00720977783203125,
0.0311431884765625,
-0.0484619140625,
0.062744140625,
0.022216796875,
-0.012420654296875,
0.024932861328125,
-0.006195068359375,
0.0195465087890625,
-0.082275390625,
0.004756927490234375,
0.02667236328125,
-0.028472900390625,
-0.03680419921875,
-0.00397491455078125,
0.00695037841796875,
-0.0003235340118408203,
-0.040283203125,
0.0246429443359375,
-0.041961669921875,
-0.00911712646484375,
-0.01082611083984375,
-0.0227508544921875,
0.0011653900146484375,
0.048675537109375,
0.001476287841796875,
0.042572021484375,
0.047332763671875,
-0.03765869140625,
0.041839599609375,
0.0214080810546875,
-0.0287017822265625,
0.048187255859375,
-0.061065673828125,
0.0191497802734375,
-0.01010894775390625,
0.0229644775390625,
-0.081298828125,
-0.01885986328125,
0.01519775390625,
-0.0408935546875,
0.040435791015625,
-0.0196685791015625,
-0.0277862548828125,
-0.061553955078125,
-0.023834228515625,
0.033294677734375,
0.051788330078125,
-0.052764892578125,
0.0281829833984375,
0.01236724853515625,
0.0299224853515625,
-0.05450439453125,
-0.0782470703125,
-0.00702667236328125,
-0.0200347900390625,
-0.046356201171875,
0.038604736328125,
0.0033435821533203125,
0.01219940185546875,
0.0149993896484375,
0.001033782958984375,
-0.0155487060546875,
-0.01282501220703125,
0.03424072265625,
0.02703857421875,
-0.0212249755859375,
-0.0027027130126953125,
-0.0260772705078125,
-0.01666259765625,
-0.0072021484375,
-0.0335693359375,
0.032623291015625,
-0.0310821533203125,
-0.021728515625,
-0.061981201171875,
0.004810333251953125,
0.04949951171875,
-0.008636474609375,
0.049713134765625,
0.06488037109375,
-0.039703369140625,
0.0081939697265625,
-0.046234130859375,
-0.0183868408203125,
-0.039154052734375,
0.025665283203125,
-0.0302581787109375,
-0.044403076171875,
0.052764892578125,
0.00732421875,
0.0014562606811523438,
0.057464599609375,
0.03326416015625,
-0.017730712890625,
0.07025146484375,
0.041839599609375,
-0.005161285400390625,
0.058746337890625,
-0.06475830078125,
0.0032558441162109375,
-0.05010986328125,
-0.013397216796875,
-0.0165557861328125,
-0.053558349609375,
-0.049957275390625,
-0.0296783447265625,
0.0214691162109375,
0.004245758056640625,
-0.03369140625,
0.048797607421875,
-0.061981201171875,
0.0248260498046875,
0.05853271484375,
0.04107666015625,
-0.00836181640625,
0.0216217041015625,
-0.006206512451171875,
-0.00334930419921875,
-0.04559326171875,
-0.01085662841796875,
0.0673828125,
0.034820556640625,
0.0516357421875,
-0.0197906494140625,
0.045867919921875,
0.01064300537109375,
0.01184844970703125,
-0.056427001953125,
0.042205810546875,
-0.0157623291015625,
-0.049407958984375,
-0.007175445556640625,
-0.027069091796875,
-0.07080078125,
0.00911712646484375,
-0.0158843994140625,
-0.046905517578125,
0.041107177734375,
0.0240631103515625,
-0.019622802734375,
0.037445068359375,
-0.055084228515625,
0.065185546875,
-0.0179443359375,
-0.03131103515625,
0.00923919677734375,
-0.0604248046875,
0.01019287109375,
-0.000014543533325195312,
-0.006931304931640625,
0.01335906982421875,
0.0208587646484375,
0.05712890625,
-0.059539794921875,
0.07025146484375,
-0.02325439453125,
0.0208587646484375,
0.05078125,
-0.0181884765625,
0.0271453857421875,
-0.0201416015625,
0.0093231201171875,
0.033172607421875,
0.0000826716423034668,
-0.036285400390625,
-0.036224365234375,
0.04443359375,
-0.06927490234375,
-0.023101806640625,
-0.03515625,
-0.0201568603515625,
0.01248931884765625,
0.01544952392578125,
0.055450439453125,
0.040008544921875,
0.006153106689453125,
0.041290283203125,
0.047393798828125,
-0.0205078125,
0.03546142578125,
-0.0177001953125,
0.0002868175506591797,
-0.0282135009765625,
0.0667724609375,
0.0308685302734375,
0.0184326171875,
0.0224609375,
0.0163116455078125,
-0.0193023681640625,
-0.0287017822265625,
-0.0254669189453125,
0.011749267578125,
-0.06207275390625,
-0.038787841796875,
-0.04864501953125,
-0.047515869140625,
-0.03167724609375,
-0.01099395751953125,
-0.0458984375,
-0.0282135009765625,
-0.0325927734375,
-0.01059722900390625,
0.0445556640625,
0.04632568359375,
-0.021392822265625,
0.038360595703125,
-0.04510498046875,
0.0146484375,
0.030517578125,
0.0297393798828125,
-0.004695892333984375,
-0.05279541015625,
-0.0264129638671875,
0.010894775390625,
-0.024322509765625,
-0.0523681640625,
0.028778076171875,
0.0203704833984375,
0.041656494140625,
0.039581298828125,
-0.011688232421875,
0.07366943359375,
-0.0193328857421875,
0.052459716796875,
0.03497314453125,
-0.04681396484375,
0.050811767578125,
-0.01288604736328125,
0.0099639892578125,
0.037841796875,
0.035369873046875,
-0.0192718505859375,
0.0010480880737304688,
-0.0572509765625,
-0.0577392578125,
0.052642822265625,
0.01340484619140625,
0.008331298828125,
0.007717132568359375,
0.04522705078125,
-0.01459503173828125,
0.0014324188232421875,
-0.06243896484375,
-0.031768798828125,
-0.038604736328125,
-0.015472412109375,
0.0026073455810546875,
-0.0152587890625,
0.007343292236328125,
-0.056304931640625,
0.0419921875,
-0.0064544677734375,
0.0548095703125,
0.01934814453125,
-0.01605224609375,
0.0012865066528320312,
-0.032470703125,
0.0130462646484375,
0.031951904296875,
-0.01514434814453125,
0.00431060791015625,
0.00772857666015625,
-0.05694580078125,
0.00762939453125,
0.005558013916015625,
-0.01140594482421875,
-0.0020904541015625,
0.03179931640625,
0.07586669921875,
-0.0010623931884765625,
-0.00091552734375,
0.058074951171875,
-0.0112152099609375,
-0.034637451171875,
-0.0308990478515625,
0.0025463104248046875,
-0.01256561279296875,
0.0284576416015625,
0.0278167724609375,
0.021942138671875,
0.00676727294921875,
-0.02325439453125,
0.02850341796875,
0.0252227783203125,
-0.041412353515625,
-0.02447509765625,
0.048675537109375,
-0.007221221923828125,
0.007656097412109375,
0.057098388671875,
-0.00679779052734375,
-0.043853759765625,
0.0782470703125,
0.0292205810546875,
0.061614990234375,
-0.020172119140625,
0.0118408203125,
0.060546875,
0.0202178955078125,
-0.0110931396484375,
0.005283355712890625,
0.0046539306640625,
-0.05328369140625,
-0.0188446044921875,
-0.049652099609375,
0.018524169921875,
0.021575927734375,
-0.054107666015625,
0.02679443359375,
-0.035980224609375,
-0.0408935546875,
0.018341064453125,
0.0014581680297851562,
-0.083251953125,
0.029693603515625,
0.01197052001953125,
0.0633544921875,
-0.05999755859375,
0.057525634765625,
0.05706787109375,
-0.04461669921875,
-0.07537841796875,
-0.0216522216796875,
-0.004199981689453125,
-0.055877685546875,
0.0643310546875,
0.031585693359375,
0.0118865966796875,
0.01302337646484375,
-0.056396484375,
-0.072265625,
0.10003662109375,
0.0225830078125,
-0.03228759765625,
0.0009760856628417969,
0.01099395751953125,
0.033660888671875,
-0.02362060546875,
0.040283203125,
0.024505615234375,
0.0274810791015625,
0.0323486328125,
-0.05902099609375,
0.00635528564453125,
-0.032684326171875,
0.025726318359375,
-0.0033588409423828125,
-0.06561279296875,
0.0782470703125,
-0.008209228515625,
-0.01010894775390625,
-0.0041046142578125,
0.050445556640625,
-0.010284423828125,
-0.0019178390502929688,
0.05816650390625,
0.06036376953125,
0.0292205810546875,
-0.0228729248046875,
0.07733154296875,
-0.005321502685546875,
0.04241943359375,
0.047637939453125,
0.0269775390625,
0.032379150390625,
0.0273590087890625,
-0.023101806640625,
0.02703857421875,
0.08343505859375,
-0.025146484375,
0.043304443359375,
0.001934051513671875,
0.0081939697265625,
-0.012939453125,
0.0007419586181640625,
-0.0374755859375,
0.038818359375,
0.016510009765625,
-0.050994873046875,
-0.003765106201171875,
0.0196380615234375,
-0.01335906982421875,
-0.0279388427734375,
-0.027374267578125,
0.0517578125,
0.005771636962890625,
-0.034637451171875,
0.06329345703125,
-0.01617431640625,
0.056396484375,
-0.0223541259765625,
-0.004192352294921875,
-0.0198822021484375,
0.031005859375,
-0.025726318359375,
-0.05511474609375,
0.01739501953125,
-0.006710052490234375,
-0.0033016204833984375,
-0.0092315673828125,
0.069580078125,
-0.0131683349609375,
-0.046173095703125,
0.021453857421875,
0.0181884765625,
0.0222320556640625,
-0.00693511962890625,
-0.07147216796875,
0.0009984970092773438,
-0.0021266937255859375,
-0.05072021484375,
0.0189208984375,
0.035125732421875,
-0.0012731552124023438,
0.03387451171875,
0.05047607421875,
-0.005268096923828125,
0.020660400390625,
-0.0133209228515625,
0.0848388671875,
-0.034942626953125,
-0.035736083984375,
-0.057037353515625,
0.0465087890625,
-0.0221099853515625,
-0.0224456787109375,
0.04425048828125,
0.0323486328125,
0.07208251953125,
-0.004352569580078125,
0.050201416015625,
-0.0251922607421875,
0.0097198486328125,
-0.017913818359375,
0.040191650390625,
-0.0548095703125,
-0.0159759521484375,
-0.0301513671875,
-0.07733154296875,
-0.0178070068359375,
0.074462890625,
-0.01166534423828125,
0.0313720703125,
0.03997802734375,
0.058013916015625,
-0.02435302734375,
-0.0149993896484375,
0.017242431640625,
0.0164947509765625,
0.01390838623046875,
0.03460693359375,
0.05224609375,
-0.058746337890625,
0.037628173828125,
-0.060394287109375,
-0.0274658203125,
-0.021697998046875,
-0.057342529296875,
-0.07293701171875,
-0.061981201171875,
-0.03875732421875,
-0.03643798828125,
-0.009613037109375,
0.060638427734375,
0.08123779296875,
-0.051910400390625,
0.004917144775390625,
-0.01025390625,
-0.0200653076171875,
-0.029266357421875,
-0.017120361328125,
0.043548583984375,
-0.0020122528076171875,
-0.0640869140625,
-0.0204925537109375,
-0.0016813278198242188,
0.02630615234375,
-0.01274871826171875,
-0.00467681884765625,
-0.01910400390625,
-0.0216522216796875,
0.045989990234375,
0.0152435302734375,
-0.03143310546875,
-0.031890869140625,
0.006519317626953125,
-0.0167694091796875,
0.0237884521484375,
0.040618896484375,
-0.0487060546875,
0.024688720703125,
0.043853759765625,
0.0439453125,
0.06256103515625,
0.005542755126953125,
0.0014696121215820312,
-0.052825927734375,
0.032684326171875,
0.003231048583984375,
0.034088134765625,
0.0206298828125,
-0.040191650390625,
0.05364990234375,
0.03533935546875,
-0.044525146484375,
-0.0550537109375,
-0.0059967041015625,
-0.091064453125,
-0.01036834716796875,
0.07568359375,
-0.02740478515625,
-0.040130615234375,
0.0233306884765625,
-0.011444091796875,
0.03912353515625,
-0.0116729736328125,
0.03204345703125,
0.035797119140625,
0.00202178955078125,
-0.0295257568359375,
-0.050323486328125,
0.026397705078125,
0.0037288665771484375,
-0.045013427734375,
-0.027191162109375,
0.024322509765625,
0.0239410400390625,
0.03216552734375,
0.04473876953125,
-0.0249786376953125,
0.007778167724609375,
0.005462646484375,
0.01416015625,
-0.01221466064453125,
-0.021881103515625,
-0.0127105712890625,
-0.01029205322265625,
-0.0216522216796875,
-0.049957275390625
]
] |
TheLastBen/Pikachu_SDXL | 2023-08-29T10:35:30.000Z | [
"diffusers",
"text-to-image",
"stable-diffusion",
"lora",
"license:creativeml-openrail-m",
"region:us",
"has_space"
] | text-to-image | TheLastBen | null | null | TheLastBen/Pikachu_SDXL | 1 | 7,504 | diffusers | 2023-08-12T15:22:27 | ---
license: creativeml-openrail-m
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: pikachu
widget:
- text: pikachu
---
### Pikachu
#### SDXL LoRA by TheLastBen
#### Prompts to start with :
closeup on pikachu on a pirate ship, cinematic, screencap, high quality, light rays, sunrays, pov, ships, 1800s
closeup on fluffy pikachu wearing a hoodie in a street in london, cinematic, screencap, high quality
---
Trained using https://github.com/TheLastBen/fast-stable-diffusion SDXL trainer.
ComfyUI seems to give better results than A1111, but that's just me.
#### Sample pictures:
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp)
.webp) | 2,995 | [
[
-0.06256103515625,
-0.0297698974609375,
0.01151275634765625,
0.05352783203125,
-0.0240020751953125,
-0.0004363059997558594,
0.00319671630859375,
-0.046051025390625,
0.1026611328125,
0.003345489501953125,
-0.05816650390625,
-0.0164947509765625,
-0.04730224609375,
0.00823211669921875,
0.0124359130859375,
0.053070068359375,
0.01136016845703125,
-0.01428985595703125,
-0.007503509521484375,
-0.00492095947265625,
-0.0030422210693359375,
-0.0283355712890625,
-0.041351318359375,
-0.045501708984375,
0.0428466796875,
0.020965576171875,
0.05535888671875,
0.0037746429443359375,
0.0285186767578125,
0.032257080078125,
-0.0195465087890625,
0.0009107589721679688,
-0.016876220703125,
0.007068634033203125,
-0.00286865234375,
-0.040496826171875,
-0.07037353515625,
-0.004791259765625,
0.0280303955078125,
0.043975830078125,
-0.0011577606201171875,
0.004428863525390625,
-0.0000909566879272461,
0.06573486328125,
-0.01580810546875,
0.00540924072265625,
-0.0070343017578125,
0.0218658447265625,
-0.006290435791015625,
0.00958251953125,
-0.00766754150390625,
-0.049468994140625,
0.00021541118621826172,
-0.0572509765625,
0.0021572113037109375,
-0.01016998291015625,
0.09906005859375,
-0.020660400390625,
-0.006587982177734375,
-0.0217742919921875,
0.0150604248046875,
0.0562744140625,
-0.028076171875,
0.0029239654541015625,
0.038787841796875,
-0.004726409912109375,
-0.0249481201171875,
-0.043487548828125,
-0.0309906005859375,
0.004032135009765625,
-0.01294708251953125,
0.0227203369140625,
-0.058074951171875,
-0.034393310546875,
0.0205078125,
0.032745361328125,
-0.03778076171875,
-0.005542755126953125,
-0.03851318359375,
-0.0183868408203125,
0.06787109375,
-0.0124969482421875,
0.05517578125,
-0.01497650146484375,
-0.037506103515625,
-0.0223846435546875,
-0.01654052734375,
0.0276641845703125,
0.0172119140625,
-0.007266998291015625,
-0.04986572265625,
0.05255126953125,
0.01371002197265625,
0.01383209228515625,
0.06494140625,
-0.02740478515625,
0.039703369140625,
-0.0175628662109375,
0.001617431640625,
0.00373077392578125,
0.07501220703125,
0.0648193359375,
0.0222625732421875,
-0.00024700164794921875,
0.01849365234375,
0.0006365776062011719,
0.00261688232421875,
-0.049835205078125,
-0.017425537109375,
0.0271148681640625,
-0.050048828125,
-0.0249481201171875,
-0.0017147064208984375,
-0.07769775390625,
-0.0233612060546875,
0.01105499267578125,
-0.0177001953125,
-0.0697021484375,
-0.042205810546875,
0.00971221923828125,
-0.0303802490234375,
0.03900146484375,
0.0328369140625,
-0.052001953125,
0.01052093505859375,
0.025482177734375,
0.047637939453125,
0.0206298828125,
-0.018402099609375,
-0.0175628662109375,
0.0006275177001953125,
-0.0264739990234375,
0.05950927734375,
-0.032440185546875,
-0.02545166015625,
-0.0426025390625,
0.01291656494140625,
-0.0061492919921875,
-0.004886627197265625,
0.046600341796875,
0.0161285400390625,
0.0150299072265625,
-0.0260162353515625,
-0.038604736328125,
-0.017364501953125,
0.01395416259765625,
-0.053985595703125,
0.053985595703125,
0.0282745361328125,
-0.070068359375,
0.0061492919921875,
-0.05450439453125,
-0.0030155181884765625,
-0.0056915283203125,
0.022552490234375,
-0.043792724609375,
0.013641357421875,
0.0013265609741210938,
0.037933349609375,
0.006378173828125,
-0.04876708984375,
-0.05035400390625,
0.00617218017578125,
0.024383544921875,
0.0249481201171875,
0.078369140625,
0.0229644775390625,
-0.0270843505859375,
-0.00794219970703125,
-0.0579833984375,
0.0180511474609375,
0.056915283203125,
0.004108428955078125,
-0.0137786865234375,
-0.034423828125,
-0.00882720947265625,
0.0196380615234375,
0.0318603515625,
-0.03997802734375,
0.03192138671875,
-0.00597381591796875,
0.0006513595581054688,
0.0265655517578125,
-0.0018291473388671875,
0.036834716796875,
-0.0309600830078125,
0.047821044921875,
-0.01537322998046875,
0.02276611328125,
-0.0102081298828125,
-0.046630859375,
-0.04205322265625,
-0.037506103515625,
0.007282257080078125,
0.03131103515625,
-0.023284912109375,
0.07672119140625,
0.0034084320068359375,
-0.06817626953125,
-0.033782958984375,
-0.00231170654296875,
0.030364990234375,
-0.00420379638671875,
0.004909515380859375,
-0.041259765625,
-0.0372314453125,
-0.037933349609375,
-0.00946044921875,
-0.0218963623046875,
0.01277923583984375,
0.032379150390625,
0.0242156982421875,
0.0026264190673828125,
0.024261474609375,
-0.0204925537109375,
-0.0105133056640625,
-0.021453857421875,
-0.01445770263671875,
0.03955078125,
0.040771484375,
0.07647705078125,
-0.061676025390625,
-0.05694580078125,
-0.0159454345703125,
-0.0672607421875,
0.0001537799835205078,
-0.004810333251953125,
-0.0294342041015625,
0.01337432861328125,
0.0085296630859375,
-0.07269287109375,
0.044281005859375,
0.04302978515625,
-0.069580078125,
0.0291748046875,
-0.04388427734375,
0.037841796875,
-0.055511474609375,
0.0122528076171875,
0.00402069091796875,
-0.014617919921875,
-0.0537109375,
0.0460205078125,
-0.007289886474609375,
0.007007598876953125,
-0.044830322265625,
0.05157470703125,
-0.047119140625,
0.00879669189453125,
-0.0010423660278320312,
0.033355712890625,
0.0131988525390625,
0.0166168212890625,
0.01419830322265625,
0.01386260986328125,
0.06005859375,
-0.02252197265625,
0.06158447265625,
0.06353759765625,
0.00905609130859375,
0.09490966796875,
-0.0562744140625,
0.0025615692138671875,
-0.0017862319946289062,
0.050445556640625,
-0.0506591796875,
-0.03265380859375,
0.0513916015625,
-0.0535888671875,
0.006092071533203125,
-0.0008826255798339844,
-0.04168701171875,
-0.042083740234375,
-0.06292724609375,
0.011566162109375,
0.0670166015625,
-0.052154541015625,
0.0321044921875,
0.019866943359375,
0.00495147705078125,
-0.0283660888671875,
-0.048370361328125,
0.01461029052734375,
-0.048736572265625,
-0.06353759765625,
0.0286712646484375,
-0.01073455810546875,
-0.01169586181640625,
-0.0011091232299804688,
0.0005602836608886719,
-0.00827789306640625,
-0.01279449462890625,
0.0295562744140625,
0.01261138916015625,
-0.007602691650390625,
-0.039093017578125,
-0.032257080078125,
0.0024662017822265625,
-0.00989532470703125,
0.02325439453125,
0.05621337890625,
-0.037567138671875,
-0.01513671875,
-0.06451416015625,
0.0229949951171875,
0.04791259765625,
0.00394439697265625,
0.049835205078125,
0.0579833984375,
-0.022125244140625,
0.032928466796875,
-0.033477783203125,
0.0148773193359375,
-0.0299530029296875,
-0.037139892578125,
-0.05255126953125,
-0.0213623046875,
0.049072265625,
0.0246124267578125,
-0.01499176025390625,
0.04119873046875,
0.03375244140625,
-0.0243682861328125,
0.06866455078125,
0.041473388671875,
-0.00901031494140625,
0.035675048828125,
-0.033233642578125,
0.0160064697265625,
-0.055389404296875,
-0.0579833984375,
-0.00312042236328125,
-0.03656005859375,
-0.039031982421875,
-0.0228118896484375,
0.01024627685546875,
0.06353759765625,
-0.008758544921875,
0.0802001953125,
-0.053070068359375,
0.021392822265625,
0.041778564453125,
0.04083251953125,
-0.0005540847778320312,
0.00702667236328125,
0.0093536376953125,
-0.0099639892578125,
-0.0199737548828125,
-0.0198822021484375,
0.04083251953125,
0.036590576171875,
0.0469970703125,
0.0258026123046875,
0.06622314453125,
-0.00495147705078125,
-0.023834228515625,
-0.03167724609375,
0.0670166015625,
-0.000995635986328125,
-0.0411376953125,
-0.007633209228515625,
0.00304412841796875,
-0.10498046875,
0.01425933837890625,
-0.0345458984375,
-0.066162109375,
0.0095062255859375,
-0.00036144256591796875,
-0.017791748046875,
0.033905029296875,
-0.05157470703125,
0.021087646484375,
-0.004924774169921875,
-0.03570556640625,
-0.02685546875,
-0.050628662109375,
0.0153045654296875,
0.0026645660400390625,
-0.019927978515625,
-0.03271484375,
-0.01261138916015625,
0.037139892578125,
-0.046600341796875,
0.0361328125,
-0.0195159912109375,
-0.00565338134765625,
0.026763916015625,
-0.0071258544921875,
0.0283203125,
0.0013580322265625,
-0.0160369873046875,
-0.0116424560546875,
0.032196044921875,
-0.068603515625,
-0.0291900634765625,
0.058990478515625,
-0.0406494140625,
-0.0149078369140625,
-0.057037353515625,
-0.004547119140625,
0.008575439453125,
0.01143646240234375,
0.043670654296875,
0.012176513671875,
-0.007328033447265625,
0.0193328857421875,
0.0577392578125,
-0.0266571044921875,
0.038238525390625,
-0.01552581787109375,
-0.0204620361328125,
-0.011383056640625,
0.0477294921875,
-0.022491455078125,
-0.0010814666748046875,
-0.0024089813232421875,
0.0265350341796875,
-0.0283203125,
0.0144805908203125,
0.0025691986083984375,
0.056976318359375,
-0.03546142578125,
-0.01629638671875,
-0.03948974609375,
-0.0181427001953125,
-0.028594970703125,
-0.03656005859375,
-0.027984619140625,
-0.040496826171875,
-0.0289459228515625,
0.030242919921875,
0.065185546875,
0.046844482421875,
-0.03594970703125,
0.0131072998046875,
-0.046722412109375,
0.03240966796875,
0.01434326171875,
0.033294677734375,
-0.0240936279296875,
-0.0229339599609375,
0.051025390625,
0.0025787353515625,
-0.0172882080078125,
-0.068603515625,
0.041839599609375,
0.0010814666748046875,
0.052947998046875,
0.0196533203125,
0.00266265869140625,
0.059539794921875,
-0.0265960693359375,
0.0556640625,
0.0848388671875,
-0.039825439453125,
0.058929443359375,
-0.0206146240234375,
0.00534820556640625,
0.0200653076171875,
0.049530029296875,
-0.0265960693359375,
-0.02880859375,
-0.08538818359375,
-0.032501220703125,
0.0291595458984375,
0.0168914794921875,
0.0234375,
0.0076751708984375,
0.026275634765625,
-0.0205230712890625,
-0.0008440017700195312,
-0.041778564453125,
-0.056884765625,
-0.0255889892578125,
-0.00276947021484375,
0.008209228515625,
-0.011383056640625,
0.0010633468627929688,
-0.03851318359375,
0.04229736328125,
0.0029544830322265625,
0.04705810546875,
0.01169586181640625,
-0.00604248046875,
-0.004932403564453125,
-0.0191497802734375,
0.05413818359375,
0.043609619140625,
-0.0164794921875,
-0.0296783447265625,
0.00949859619140625,
-0.042449951171875,
0.00787353515625,
-0.00588226318359375,
0.0117950439453125,
0.022979736328125,
0.024871826171875,
0.052215576171875,
0.0205078125,
-0.03350830078125,
0.0526123046875,
-0.022216796875,
-0.020904541015625,
-0.03350830078125,
0.0126953125,
0.0091094970703125,
0.04052734375,
0.0099029541015625,
-0.001598358154296875,
-0.01367950439453125,
-0.042205810546875,
0.0306549072265625,
0.024139404296875,
-0.01039886474609375,
-0.03173828125,
0.039764404296875,
-0.006267547607421875,
-0.01450347900390625,
0.0164947509765625,
-0.0267181396484375,
-0.038330078125,
0.0543212890625,
0.0291595458984375,
0.02813720703125,
-0.03302001953125,
0.031768798828125,
0.060821533203125,
-0.0205535888671875,
0.007068634033203125,
0.0195770263671875,
0.0111083984375,
-0.0071258544921875,
0.00701904296875,
-0.03253173828125,
-0.01438140869140625,
0.0233612060546875,
-0.03790283203125,
0.0426025390625,
-0.05279541015625,
-0.0165557861328125,
0.006748199462890625,
0.039031982421875,
-0.03228759765625,
0.0214385986328125,
0.01454925537109375,
0.081298828125,
-0.042388916015625,
0.0509033203125,
0.04345703125,
-0.04620361328125,
-0.053131103515625,
-0.0008015632629394531,
0.0189971923828125,
-0.091552734375,
0.025115966796875,
-0.011444091796875,
0.019989013671875,
-0.0233001708984375,
-0.04656982421875,
-0.04974365234375,
0.0882568359375,
0.00612640380859375,
-0.01873779296875,
-0.0032501220703125,
-0.04400634765625,
0.033111572265625,
-0.0256195068359375,
0.036956787109375,
0.0511474609375,
0.038055419921875,
0.031280517578125,
-0.06585693359375,
0.0138397216796875,
-0.0426025390625,
-0.007465362548828125,
-0.0018291473388671875,
-0.08648681640625,
0.04205322265625,
-0.0279998779296875,
0.003780364990234375,
0.037689208984375,
0.0667724609375,
0.055694580078125,
-0.0023746490478515625,
0.05645751953125,
0.0438232421875,
0.038330078125,
-0.012969970703125,
0.0802001953125,
-0.01552581787109375,
0.03265380859375,
0.0299530029296875,
0.0240478515625,
0.04571533203125,
0.031585693359375,
-0.04937744140625,
0.0294189453125,
0.0716552734375,
-0.037689208984375,
0.0343017578125,
0.024658203125,
-0.03558349609375,
0.004024505615234375,
-0.0099639892578125,
-0.0406494140625,
0.0208282470703125,
0.00826263427734375,
-0.0011014938354492188,
-0.0291900634765625,
0.0026836395263671875,
0.035430908203125,
0.020538330078125,
-0.035003662109375,
0.029052734375,
-0.004779815673828125,
-0.04083251953125,
0.040496826171875,
-0.0177154541015625,
0.07977294921875,
-0.03472900390625,
0.01007080078125,
-0.044769287109375,
0.00824737548828125,
-0.041778564453125,
-0.08831787109375,
0.016082763671875,
0.005130767822265625,
0.009124755859375,
-0.02435302734375,
0.052764892578125,
-0.0245513916015625,
-0.039825439453125,
0.024993896484375,
0.0009222030639648438,
0.0633544921875,
0.0259246826171875,
-0.1002197265625,
0.029815673828125,
0.0089263916015625,
-0.04083251953125,
0.02490234375,
0.0233917236328125,
0.019927978515625,
0.06011962890625,
0.0177001953125,
0.0262451171875,
0.0235748291015625,
-0.02130126953125,
0.033477783203125,
-0.0183868408203125,
-0.01052093505859375,
-0.048187255859375,
0.03472900390625,
-0.041839599609375,
-0.0155029296875,
0.06494140625,
0.0165252685546875,
0.020660400390625,
-0.00872039794921875,
0.03955078125,
-0.044769287109375,
0.043548583984375,
-0.01457977294921875,
0.055023193359375,
-0.062103271484375,
-0.03204345703125,
-0.025238037109375,
-0.0504150390625,
-0.01629638671875,
0.06866455078125,
0.01255035400390625,
-0.002178192138671875,
0.0308685302734375,
0.0360107421875,
-0.0077972412109375,
-0.021759033203125,
-0.0007615089416503906,
-0.022064208984375,
0.01580810546875,
0.07122802734375,
0.046722412109375,
-0.0592041015625,
0.0169219970703125,
-0.055023193359375,
-0.0158233642578125,
-0.0264892578125,
-0.0628662109375,
-0.08282470703125,
-0.050567626953125,
-0.050445556640625,
-0.058624267578125,
-0.03363037109375,
0.07000732421875,
0.06414794921875,
-0.054901123046875,
-0.01023101806640625,
0.01336669921875,
0.007602691650390625,
0.011474609375,
-0.01177978515625,
0.0283660888671875,
0.00113677978515625,
-0.0809326171875,
0.0024318695068359375,
0.0029468536376953125,
0.039276123046875,
0.0142669677734375,
-0.0298919677734375,
-0.0199737548828125,
-0.0022563934326171875,
0.038055419921875,
0.0609130859375,
-0.049560546875,
-0.008148193359375,
-0.0020885467529296875,
-0.0181121826171875,
0.0085906982421875,
0.01041412353515625,
-0.0214691162109375,
-0.002696990966796875,
0.04095458984375,
0.0139007568359375,
0.050048828125,
0.00238037109375,
0.0079345703125,
-0.0316162109375,
0.024383544921875,
-0.01332855224609375,
0.037353515625,
0.0155029296875,
-0.0269012451171875,
0.072998046875,
0.02923583984375,
-0.0183258056640625,
-0.06298828125,
-0.017730712890625,
-0.09844970703125,
-0.0140228271484375,
0.0687255859375,
-0.0171661376953125,
-0.0307464599609375,
0.03192138671875,
-0.025482177734375,
0.0032405853271484375,
-0.0207977294921875,
0.01441192626953125,
0.0389404296875,
-0.0562744140625,
-0.0172576904296875,
-0.01099395751953125,
0.016143798828125,
0.021392822265625,
-0.05841064453125,
-0.031219482421875,
0.012359619140625,
0.0229644775390625,
0.050201416015625,
0.050384521484375,
-0.025146484375,
0.0261383056640625,
0.016632080078125,
0.01285552978515625,
0.024444580078125,
0.0272369384765625,
-0.0325927734375,
-0.01113128662109375,
-0.0206298828125,
-0.027923583984375
]
] |
facebook/esm2_t36_3B_UR50D | 2022-12-01T20:22:22.000Z | [
"transformers",
"pytorch",
"tf",
"esm",
"fill-mask",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | fill-mask | facebook | null | null | facebook/esm2_t36_3B_UR50D | 10 | 7,500 | transformers | 2022-10-13T12:38:30 | ---
license: mit
widget:
- text: "MQIFVKTLTGKTITLEVEPS<mask>TIENVKAKIQDKEGIPPDQQRLIFAGKQLEDGRTLSDYNIQKESTLHLVLRLRGG"
---
## ESM-2
ESM-2 is a state-of-the-art protein model trained on a masked language modelling objective. It is suitable for fine-tuning on a wide range of tasks that take protein sequences as input. For detailed information on the model architecture and training data, please refer to the [accompanying paper](https://www.biorxiv.org/content/10.1101/2022.07.20.500902v2). You may also be interested in some demo notebooks ([PyTorch](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/protein_language_modeling.ipynb), [TensorFlow](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/protein_language_modeling-tf.ipynb)) which demonstrate how to fine-tune ESM-2 models on your tasks of interest.
Several ESM-2 checkpoints are available in the Hub with varying sizes. Larger sizes generally have somewhat better accuracy, but require much more memory and time to train:
| Checkpoint name | Num layers | Num parameters |
|------------------------------|----|----------|
| [esm2_t48_15B_UR50D](https://huggingface.co/facebook/esm2_t48_15B_UR50D) | 48 | 15B |
| [esm2_t36_3B_UR50D](https://huggingface.co/facebook/esm2_t36_3B_UR50D) | 36 | 3B |
| [esm2_t33_650M_UR50D](https://huggingface.co/facebook/esm2_t33_650M_UR50D) | 33 | 650M |
| [esm2_t30_150M_UR50D](https://huggingface.co/facebook/esm2_t30_150M_UR50D) | 30 | 150M |
| [esm2_t12_35M_UR50D](https://huggingface.co/facebook/esm2_t12_35M_UR50D) | 12 | 35M |
| [esm2_t6_8M_UR50D](https://huggingface.co/facebook/esm2_t6_8M_UR50D) | 6 | 8M | | 1,705 | [
[
-0.029815673828125,
-0.041015625,
0.0237884521484375,
0.0173187255859375,
-0.014892578125,
0.0049896240234375,
0.00994873046875,
-0.03546142578125,
0.0181121826171875,
0.0285186767578125,
-0.056793212890625,
-0.03656005859375,
-0.064208984375,
0.005741119384765625,
-0.013885498046875,
0.0743408203125,
0.0011768341064453125,
0.0188751220703125,
-0.023895263671875,
-0.00672149658203125,
-0.0071868896484375,
-0.0170440673828125,
-0.05670166015625,
-0.050445556640625,
0.0230560302734375,
0.033294677734375,
0.0191802978515625,
0.046905517578125,
0.0333251953125,
0.016204833984375,
-0.034912109375,
0.02490234375,
-0.03607177734375,
0.01320648193359375,
-0.00881195068359375,
-0.0305938720703125,
-0.055877685546875,
-0.01080322265625,
0.031890869140625,
0.03692626953125,
0.00409698486328125,
0.03448486328125,
0.013763427734375,
0.06524658203125,
-0.0227813720703125,
0.0138397216796875,
-0.0295867919921875,
0.0201873779296875,
-0.0207366943359375,
-0.0020313262939453125,
-0.0264739990234375,
0.00366973876953125,
0.004489898681640625,
-0.0245513916015625,
0.01410675048828125,
0.01074981689453125,
0.09185791015625,
0.0157470703125,
-0.044219970703125,
-0.016021728515625,
-0.0306549072265625,
0.05950927734375,
-0.034423828125,
0.030426025390625,
0.052001953125,
0.0215301513671875,
-0.0214691162109375,
-0.053619384765625,
-0.002605438232421875,
0.021453857421875,
-0.0012607574462890625,
0.028076171875,
-0.018646240234375,
0.01357269287109375,
0.039581298828125,
0.0223236083984375,
-0.0679931640625,
0.01380157470703125,
-0.04443359375,
-0.01549530029296875,
0.0406494140625,
0.01465606689453125,
0.02362060546875,
0.001995086669921875,
-0.032958984375,
0.01082611083984375,
-0.039947509765625,
0.0036773681640625,
0.02056884765625,
-0.003032684326171875,
-0.015838623046875,
0.043548583984375,
-0.033111572265625,
0.053955078125,
0.00534820556640625,
-0.01172637939453125,
0.034088134765625,
-0.0039215087890625,
0.002483367919921875,
-0.0355224609375,
0.038818359375,
0.0509033203125,
-0.006389617919921875,
-0.007747650146484375,
-0.031097412109375,
-0.007965087890625,
0.003917694091796875,
-0.09149169921875,
-0.0121307373046875,
0.04443359375,
-0.03631591796875,
-0.0167236328125,
0.00991058349609375,
-0.056884765625,
-0.0008487701416015625,
-0.0178985595703125,
0.0267333984375,
-0.037384033203125,
-0.0174407958984375,
0.01221466064453125,
-0.033660888671875,
0.02471923828125,
0.0158233642578125,
-0.061431884765625,
0.044921875,
0.050384521484375,
0.08306884765625,
-0.0030727386474609375,
-0.0159912109375,
-0.031280517578125,
0.02044677734375,
-0.01015472412109375,
0.061859130859375,
-0.0154266357421875,
-0.003803253173828125,
0.002227783203125,
0.0228118896484375,
-0.004528045654296875,
-0.03857421875,
0.027679443359375,
-0.0217742919921875,
0.0111236572265625,
-0.03057861328125,
-0.056640625,
-0.0291900634765625,
0.00867462158203125,
-0.0303192138671875,
0.10418701171875,
0.01529693603515625,
-0.0408935546875,
0.0075225830078125,
-0.041107177734375,
-0.025146484375,
-0.0007357597351074219,
-0.01081085205078125,
-0.05352783203125,
0.0085296630859375,
-0.01369476318359375,
0.0305023193359375,
-0.0233612060546875,
0.002254486083984375,
-0.02398681640625,
-0.025787353515625,
0.0053863525390625,
0.032257080078125,
0.05029296875,
0.03363037109375,
-0.042877197265625,
-0.022613525390625,
-0.06646728515625,
0.018707275390625,
0.0158233642578125,
-0.0197906494140625,
0.0256195068359375,
0.0021610260009765625,
0.0164642333984375,
0.045745849609375,
0.018157958984375,
-0.03411865234375,
0.005626678466796875,
-0.0230560302734375,
0.04656982421875,
0.032958984375,
0.0015201568603515625,
0.020782470703125,
-0.04913330078125,
0.0296478271484375,
0.003185272216796875,
0.006267547607421875,
-0.007755279541015625,
-0.061187744140625,
-0.06744384765625,
-0.028350830078125,
-0.006290435791015625,
0.04888916015625,
-0.0197906494140625,
0.0550537109375,
0.0122833251953125,
-0.042755126953125,
-0.0269012451171875,
0.00824737548828125,
0.040374755859375,
0.018463134765625,
0.033782958984375,
-0.01462554931640625,
-0.057373046875,
-0.08612060546875,
-0.0274505615234375,
0.0015239715576171875,
-0.022552490234375,
0.0152130126953125,
0.0595703125,
-0.026641845703125,
0.0546875,
-0.02667236328125,
-0.0219879150390625,
-0.01270294189453125,
0.0086517333984375,
0.00937652587890625,
0.047821044921875,
0.0462646484375,
-0.032196044921875,
-0.031158447265625,
-0.008636474609375,
-0.05731201171875,
-0.015289306640625,
0.01125335693359375,
-0.0035800933837890625,
0.0113677978515625,
0.045196533203125,
-0.034423828125,
0.0084991455078125,
0.051055908203125,
-0.04669189453125,
0.009918212890625,
-0.0117645263671875,
-0.0023593902587890625,
-0.09649658203125,
0.0152740478515625,
0.00007420778274536133,
-0.03424072265625,
-0.046783447265625,
0.007625579833984375,
0.01256561279296875,
-0.015167236328125,
-0.0416259765625,
0.048370361328125,
-0.0540771484375,
-0.0234527587890625,
-0.0274505615234375,
-0.002132415771484375,
0.018707275390625,
0.03668212890625,
0.0012788772583007812,
0.03265380859375,
0.054534912109375,
-0.0256195068359375,
0.0040130615234375,
0.0270538330078125,
-0.0220489501953125,
0.031219482421875,
-0.06622314453125,
0.03704833984375,
-0.0192413330078125,
0.0238800048828125,
-0.07354736328125,
-0.0330810546875,
0.01125335693359375,
-0.03851318359375,
0.03802490234375,
-0.0248870849609375,
-0.0335693359375,
-0.03631591796875,
-0.032928466796875,
0.0127105712890625,
0.05633544921875,
-0.0262298583984375,
0.032196044921875,
0.041229248046875,
-0.002796173095703125,
-0.0264739990234375,
-0.07135009765625,
-0.008544921875,
-0.0088958740234375,
-0.047698974609375,
0.034454345703125,
0.0021305084228515625,
0.01171112060546875,
-0.012451171875,
-0.01016998291015625,
0.00807952880859375,
0.0018491744995117188,
0.0472412109375,
-0.0031871795654296875,
0.00846099853515625,
-0.013824462890625,
0.0275115966796875,
-0.02154541015625,
-0.0131072998046875,
-0.01898193359375,
0.048248291015625,
-0.03558349609375,
-0.01416778564453125,
-0.04949951171875,
0.03375244140625,
0.050628662109375,
-0.00814056396484375,
0.065673828125,
0.060302734375,
-0.06109619140625,
-0.0086517333984375,
-0.04327392578125,
-0.032623291015625,
-0.031341552734375,
0.059906005859375,
-0.04376220703125,
-0.0771484375,
0.05670166015625,
-0.01534271240234375,
0.0015058517456054688,
0.04052734375,
0.047119140625,
-0.0189208984375,
0.0902099609375,
0.027740478515625,
0.0260772705078125,
0.029144287109375,
-0.03466796875,
-0.006317138671875,
-0.07073974609375,
-0.0582275390625,
-0.044403076171875,
-0.03399658203125,
-0.03118896484375,
-0.03631591796875,
0.00911712646484375,
0.04168701171875,
-0.039642333984375,
0.0491943359375,
-0.025054931640625,
0.033843994140625,
0.0179901123046875,
0.0227203369140625,
-0.0136566162109375,
0.01508331298828125,
-0.006275177001953125,
0.003032684326171875,
-0.0616455078125,
-0.04449462890625,
0.0653076171875,
0.06353759765625,
0.0335693359375,
0.00972747802734375,
0.04510498046875,
0.00743865966796875,
-0.006877899169921875,
-0.060516357421875,
0.035308837890625,
-0.0112152099609375,
-0.0589599609375,
-0.006412506103515625,
-0.0142669677734375,
-0.049652099609375,
0.01009368896484375,
-0.0161590576171875,
-0.06939697265625,
-0.0020236968994140625,
0.01318359375,
-0.0159454345703125,
0.0242156982421875,
-0.03961181640625,
0.046661376953125,
0.0008668899536132812,
-0.02325439453125,
-0.0089111328125,
-0.061187744140625,
-0.0017337799072265625,
0.0001264810562133789,
0.0059967041015625,
-0.0296173095703125,
-0.0128631591796875,
0.07513427734375,
-0.041748046875,
0.0552978515625,
-0.01238250732421875,
0.0255889892578125,
0.0203857421875,
0.0025234222412109375,
0.066162109375,
0.00704193115234375,
-0.0131072998046875,
0.022705078125,
0.00841522216796875,
-0.0628662109375,
-0.0172119140625,
0.037384033203125,
-0.0740966796875,
-0.01154327392578125,
-0.040679931640625,
-0.0243072509765625,
-0.0138092041015625,
0.017547607421875,
0.053253173828125,
0.034423828125,
-0.00103759765625,
0.0256195068359375,
0.045562744140625,
-0.01849365234375,
0.0187835693359375,
0.053192138671875,
-0.0161285400390625,
-0.039520263671875,
0.0452880859375,
0.0186920166015625,
0.0248870849609375,
0.024993896484375,
-0.0092315673828125,
-0.031707763671875,
-0.043304443359375,
-0.030548095703125,
0.0194091796875,
-0.03564453125,
-0.031463623046875,
-0.0804443359375,
-0.0231781005859375,
-0.0273590087890625,
-0.01009368896484375,
-0.057373046875,
-0.033233642578125,
-0.0161590576171875,
-0.020355224609375,
0.043914794921875,
0.046875,
-0.0206146240234375,
0.0167083740234375,
-0.045501708984375,
0.01413726806640625,
0.0126953125,
0.029022216796875,
-0.032501220703125,
-0.06964111328125,
-0.0136871337890625,
-0.002960205078125,
-0.0203399658203125,
-0.07275390625,
0.0171356201171875,
0.038543701171875,
0.032562255859375,
0.032562255859375,
-0.0296630859375,
0.027008056640625,
-0.0310516357421875,
0.05181884765625,
0.027008056640625,
-0.048431396484375,
0.056793212890625,
-0.03302001953125,
0.0208282470703125,
0.04730224609375,
0.025848388671875,
-0.0489501953125,
-0.0305938720703125,
-0.0389404296875,
-0.059173583984375,
0.06439208984375,
0.0241241455078125,
-0.0011682510375976562,
-0.0103759765625,
0.032684326171875,
0.007488250732421875,
0.004566192626953125,
-0.03216552734375,
-0.0367431640625,
0.00424957275390625,
-0.00563812255859375,
0.01409149169921875,
-0.052734375,
-0.01340484619140625,
-0.0225372314453125,
0.075439453125,
-0.012176513671875,
0.038299560546875,
0.0015125274658203125,
-0.00215911865234375,
-0.0300750732421875,
-0.01210784912109375,
0.054718017578125,
0.03839111328125,
-0.038604736328125,
0.007297515869140625,
0.0272674560546875,
-0.03179931640625,
-0.0032291412353515625,
0.006805419921875,
-0.029541015625,
0.003498077392578125,
0.0164794921875,
0.06451416015625,
0.00649261474609375,
-0.0341796875,
0.039642333984375,
0.015045166015625,
-0.03277587890625,
-0.0152740478515625,
-0.0084075927734375,
0.0231170654296875,
0.031646728515625,
0.0161590576171875,
0.01364898681640625,
0.01053619384765625,
-0.043487548828125,
0.0322265625,
0.019134521484375,
-0.04583740234375,
-0.028411865234375,
0.05413818359375,
0.012939453125,
-0.0282745361328125,
0.0513916015625,
-0.037078857421875,
-0.046234130859375,
0.0589599609375,
0.060821533203125,
0.05908203125,
-0.018951416015625,
0.0161590576171875,
0.06768798828125,
0.0206298828125,
-0.0282440185546875,
0.04473876953125,
0.0280914306640625,
-0.043914794921875,
-0.00982666015625,
-0.06500244140625,
-0.0035610198974609375,
0.034423828125,
-0.06707763671875,
0.0379638671875,
-0.0285797119140625,
-0.019256591796875,
-0.007476806640625,
0.0124969482421875,
-0.0576171875,
0.01177215576171875,
0.006801605224609375,
0.08245849609375,
-0.08135986328125,
0.066162109375,
0.072509765625,
-0.01837158203125,
-0.0347900390625,
-0.037567138671875,
0.029998779296875,
-0.06561279296875,
0.015411376953125,
0.0281524658203125,
0.01261138916015625,
0.005237579345703125,
-0.0252685546875,
-0.06549072265625,
0.10992431640625,
0.0168304443359375,
-0.06646728515625,
0.01236724853515625,
0.003353118896484375,
0.0377197265625,
-0.0205078125,
0.0286865234375,
0.035308837890625,
0.015777587890625,
0.00902557373046875,
-0.047576904296875,
0.00858306884765625,
-0.0372314453125,
0.01367950439453125,
0.0117034912109375,
-0.08514404296875,
0.05303955078125,
-0.019866943359375,
-0.0027141571044921875,
0.031158447265625,
0.0447998046875,
0.043853759765625,
0.0303192138671875,
0.021942138671875,
0.057708740234375,
0.052093505859375,
-0.0233917236328125,
0.06097412109375,
-0.03570556640625,
0.06524658203125,
0.06878662109375,
-0.0012903213500976562,
0.04241943359375,
0.03948974609375,
-0.0217742919921875,
0.01959228515625,
0.07720947265625,
-0.0194854736328125,
0.0323486328125,
0.022796630859375,
-0.004322052001953125,
-0.0251007080078125,
-0.005161285400390625,
-0.046173095703125,
0.0125274658203125,
0.0169830322265625,
-0.0270538330078125,
-0.01393890380859375,
-0.003955841064453125,
0.0099639892578125,
-0.0123291015625,
-0.00127410888671875,
0.052276611328125,
0.01654052734375,
-0.0303192138671875,
0.0220489501953125,
0.0205841064453125,
0.0306549072265625,
-0.041656494140625,
0.003505706787109375,
-0.033538818359375,
0.00661468505859375,
-0.026031494140625,
-0.047515869140625,
0.0237274169921875,
0.004791259765625,
-0.0178070068359375,
-0.022125244140625,
0.058624267578125,
-0.037017822265625,
-0.037567138671875,
0.033447265625,
0.035858154296875,
0.035369873046875,
-0.0035858154296875,
-0.0738525390625,
0.015716552734375,
-0.016143798828125,
-0.03857421875,
0.03363037109375,
0.00844573974609375,
0.0238800048828125,
0.0458984375,
0.01568603515625,
-0.0125732421875,
-0.0089874267578125,
0.004779815673828125,
0.050628662109375,
-0.04034423828125,
-0.031829833984375,
-0.050537109375,
0.037506103515625,
-0.0048675537109375,
-0.0294189453125,
0.049041748046875,
0.07989501953125,
0.06219482421875,
-0.0169677734375,
0.039520263671875,
-0.014373779296875,
0.039276123046875,
-0.03558349609375,
0.043731689453125,
-0.05352783203125,
-0.008148193359375,
-0.00048160552978515625,
-0.06658935546875,
-0.01261138916015625,
0.048614501953125,
0.006649017333984375,
0.01154327392578125,
0.044219970703125,
0.0784912109375,
0.010772705078125,
-0.00738525390625,
0.012939453125,
0.01055145263671875,
0.0096282958984375,
0.05462646484375,
0.05084228515625,
-0.0694580078125,
0.0123291015625,
-0.01837158203125,
-0.0291900634765625,
-0.0299224853515625,
-0.0445556640625,
-0.08013916015625,
-0.053802490234375,
-0.04180908203125,
-0.05511474609375,
0.0207366943359375,
0.0806884765625,
0.0772705078125,
-0.077392578125,
-0.00991058349609375,
-0.013275146484375,
-0.0201416015625,
-0.0253448486328125,
-0.01009368896484375,
0.0174407958984375,
-0.012359619140625,
-0.0631103515625,
0.0235748291015625,
0.039276123046875,
0.0179901123046875,
0.01502227783203125,
-0.0333251953125,
-0.01995849609375,
0.0005397796630859375,
0.0509033203125,
0.027557373046875,
-0.042724609375,
-0.0261688232421875,
0.0014963150024414062,
-0.019134521484375,
-0.0065765380859375,
0.0283050537109375,
-0.00672149658203125,
0.0223846435546875,
0.047088623046875,
0.0296173095703125,
0.0726318359375,
-0.014739990234375,
0.0293121337890625,
-0.0489501953125,
0.0205078125,
0.00817108154296875,
0.0240478515625,
0.00638580322265625,
-0.0107269287109375,
0.048858642578125,
0.0302886962890625,
-0.04071044921875,
-0.0576171875,
0.027099609375,
-0.0877685546875,
-0.0222930908203125,
0.10821533203125,
0.0018091201782226562,
-0.0008492469787597656,
-0.00127410888671875,
-0.0042877197265625,
0.029754638671875,
-0.018341064453125,
0.04443359375,
0.052490234375,
-0.0165557861328125,
-0.0011739730834960938,
-0.04718017578125,
0.0552978515625,
0.0389404296875,
-0.05621337890625,
-0.0352783203125,
0.007537841796875,
0.0396728515625,
-0.01088714599609375,
0.045074462890625,
-0.02947998046875,
0.0163116455078125,
0.0076904296875,
-0.0007152557373046875,
-0.0217132568359375,
-0.0279083251953125,
-0.0216064453125,
0.00305938720703125,
0.0026397705078125,
-0.0196990966796875
]
] |
speechlessai/speechless-llama2-dolphin-orca-platypus-13b | 2023-09-29T20:18:00.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"facebook",
"meta",
"llama-2",
"en",
"dataset:ehartford/dolphin",
"dataset:Open-Orca/OpenOrca",
"dataset:garage-bAInd/Open-Platypus",
"arxiv:2307.09288",
"text-generation-inference",
"region:us"
] | text-generation | speechlessai | null | null | speechlessai/speechless-llama2-dolphin-orca-platypus-13b | 1 | 7,500 | transformers | 2023-09-16T09:08:57 | ---
extra_gated_heading: Access Llama 2 on Hugging Face
extra_gated_description: >-
This is a form to enable access to Llama 2 on Hugging Face after you have been
granted access from Meta. Please visit the [Meta website](https://ai.meta.com/resources/models-and-libraries/llama-downloads) and accept our
license terms and acceptable use policy before submitting this form. Requests
will be processed in 1-2 days.
extra_gated_prompt: "**Your Hugging Face account email address MUST match the email you provide on the Meta website, or your request will not be approved.**"
extra_gated_button_content: Submit
extra_gated_fields:
I agree to share my name, email address and username with Meta and confirm that I have already been granted download access on the Meta website: checkbox
language:
- en
datasets:
- ehartford/dolphin
- Open-Orca/OpenOrca
- garage-bAInd/Open-Platypus
library_name: transformers
pipeline_tag: text-generation
inference: false
tags:
- facebook
- meta
- pytorch
- llama
- llama-2
---
<p><h1> speechless-llama2-dolphin-orca-platypus-13b </h1></p>
Fine-tune the meta-llama/Llama-2-13b-hf with Dolphin (2% GPT4), Orca (2% GPT4) and Platypus (40%) datasets.
| Metric | Value |
| --- | --- |
| ARC | 59.64 |
| HellaSwag | 82.65 |
| MMLU | 57.90 |
| TruthfulQA | 43.44 |
| Average | 60.91 |
# **Llama 2**
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 13B pretrained model, converted for the Hugging Face Transformers format. Links to other models can be found in the index at the bottom.
## Model Details
*Note: Use of this model is governed by the Meta license. 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.*
Meta developed and publicly released the Llama 2 family of large language models (LLMs), a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama-2-Chat, are optimized for dialogue use cases. Llama-2-Chat models outperform open-source chat models on most benchmarks we tested, and in our human evaluations for helpfulness and safety, are on par with some popular closed-source models like ChatGPT and PaLM.
**Model Developers** Meta
**Variations** Llama 2 comes in a range of parameter sizes — 7B, 13B, and 70B — as well as pretrained and fine-tuned variations.
**Input** Models input text only.
**Output** Models generate text only.
**Model Architecture** Llama 2 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.
||Training Data|Params|Content Length|GQA|Tokens|LR|
|---|---|---|---|---|---|---|
|Llama 2|*A new mix of publicly available online data*|7B|4k|✗|2.0T|3.0 x 10<sup>-4</sup>|
|Llama 2|*A new mix of publicly available online data*|13B|4k|✗|2.0T|3.0 x 10<sup>-4</sup>|
|Llama 2|*A new mix of publicly available online data*|70B|4k|✔|2.0T|1.5 x 10<sup>-4</sup>|
*Llama 2 family of models.* Token counts refer to pretraining data only. All models are trained with a global batch-size of 4M tokens. Bigger models - 70B -- use Grouped-Query Attention (GQA) for improved inference scalability.
**Model Dates** Llama 2 was trained between January 2023 and July 2023.
**Status** This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.
**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** ["Llama-2: Open Foundation and Fine-tuned Chat Models"](arxiv.org/abs/2307.09288)
## Intended Use
**Intended Use Cases** Llama 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
To get the expected features and performance for the chat versions, a specific formatting needs to be followed, including the `INST` and `<<SYS>>` tags, `BOS` and `EOS` tokens, and the whitespaces and breaklines in between (we recommend calling `strip()` on inputs to avoid double-spaces). See our reference code in github for details: [`chat_completion`](https://github.com/facebookresearch/llama/blob/main/llama/generation.py#L212).
**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 Llama 2.
## Hardware and Software
**Training Factors** We used custom training libraries, Meta's Research Super Cluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.
**Carbon Footprint** Pretraining utilized a cumulative 3.3M GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 539 tCO2eq, 100% of which were offset by Meta’s sustainability program.
||Time (GPU hours)|Power Consumption (W)|Carbon Emitted(tCO<sub>2</sub>eq)|
|---|---|---|---|
|Llama 2 7B|184320|400|31.22|
|Llama 2 13B|368640|400|62.44|
|Llama 2 70B|1720320|400|291.42|
|Total|3311616||539.00|
**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** Llama 2 was pretrained on 2 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over one million new human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.
**Data Freshness** The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023.
## Evaluation Results
In this section, we report the results for the Llama 1 and Llama 2 models on standard academic benchmarks.For all the evaluations, we use our internal evaluations library.
|Model|Size|Code|Commonsense Reasoning|World Knowledge|Reading Comprehension|Math|MMLU|BBH|AGI Eval|
|---|---|---|---|---|---|---|---|---|---|
|Llama 1|7B|14.1|60.8|46.2|58.5|6.95|35.1|30.3|23.9|
|Llama 1|13B|18.9|66.1|52.6|62.3|10.9|46.9|37.0|33.9|
|Llama 1|33B|26.0|70.0|58.4|67.6|21.4|57.8|39.8|41.7|
|Llama 1|65B|30.7|70.7|60.5|68.6|30.8|63.4|43.5|47.6|
|Llama 2|7B|16.8|63.9|48.9|61.3|14.6|45.3|32.6|29.3|
|Llama 2|13B|24.5|66.9|55.4|65.8|28.7|54.8|39.4|39.1|
|Llama 2|70B|**37.5**|**71.9**|**63.6**|**69.4**|**35.2**|**68.9**|**51.2**|**54.2**|
**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
Llama 2 is 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’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, 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/](https://ai.meta.com/llama/responsible-use-guide)
## Reporting Issues
Please report any software “bug,” or other problems with the models through one of the following means:
- Reporting issues with the model: [github.com/facebookresearch/llama](http://github.com/facebookresearch/llama)
- Reporting problematic content generated by the model: [developers.facebook.com/llama_output_feedback](http://developers.facebook.com/llama_output_feedback)
- Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)
## Llama Model Index
|Model|Llama2|Llama2-hf|Llama2-chat|Llama2-chat-hf|
|---|---|---|---|---|
|7B| [Link](https://huggingface.co/llamaste/Llama-2-7b) | [Link](https://huggingface.co/llamaste/Llama-2-7b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat-hf)|
|13B| [Link](https://huggingface.co/llamaste/Llama-2-13b) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-13b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf)|
|70B| [Link](https://huggingface.co/llamaste/Llama-2-70b) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-70b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf)|
| 10,772 | [
[
-0.0172271728515625,
-0.051483154296875,
0.0283966064453125,
0.0145416259765625,
-0.0298919677734375,
0.018310546875,
-0.004688262939453125,
-0.0576171875,
0.005970001220703125,
0.0208587646484375,
-0.052459716796875,
-0.0406494140625,
-0.05157470703125,
0.00923919677734375,
-0.012451171875,
0.078857421875,
0.0022106170654296875,
-0.0204925537109375,
-0.005275726318359375,
0.004032135009765625,
-0.03558349609375,
-0.029327392578125,
-0.040283203125,
-0.034515380859375,
0.029541015625,
0.0347900390625,
0.0458984375,
0.048828125,
0.040496826171875,
0.0177459716796875,
-0.02056884765625,
0.01934814453125,
-0.053863525390625,
-0.021026611328125,
0.00821685791015625,
-0.03399658203125,
-0.052337646484375,
0.01105499267578125,
0.02703857421875,
0.016937255859375,
-0.0208740234375,
0.037933349609375,
0.00812530517578125,
0.03875732421875,
-0.039886474609375,
0.015838623046875,
-0.0537109375,
0.0016574859619140625,
-0.016693115234375,
-0.00583648681640625,
-0.01419830322265625,
-0.0196533203125,
-0.012298583984375,
-0.06549072265625,
-0.00998687744140625,
0.005619049072265625,
0.08038330078125,
0.04840087890625,
-0.03302001953125,
-0.01091766357421875,
-0.019378662109375,
0.06890869140625,
-0.062744140625,
0.007701873779296875,
0.04302978515625,
0.0169219970703125,
-0.018829345703125,
-0.056121826171875,
-0.04998779296875,
-0.01043701171875,
0.00567626953125,
0.029541015625,
-0.03009033203125,
-0.0015888214111328125,
0.00983428955078125,
0.029632568359375,
-0.041656494140625,
0.043792724609375,
-0.035186767578125,
-0.0124359130859375,
0.07757568359375,
0.0174102783203125,
0.00214385986328125,
-0.0033550262451171875,
-0.035430908203125,
-0.020904541015625,
-0.060028076171875,
0.01355743408203125,
0.038360595703125,
-0.00122833251953125,
-0.034912109375,
0.04864501953125,
-0.0290374755859375,
0.0214080810546875,
0.0013628005981445312,
-0.041259765625,
0.03857421875,
-0.0367431640625,
-0.01885986328125,
-0.0099945068359375,
0.06951904296875,
0.054443359375,
0.01081085205078125,
0.0098724365234375,
-0.003612518310546875,
0.00655364990234375,
-0.0028896331787109375,
-0.059661865234375,
-0.004055023193359375,
0.0160980224609375,
-0.029937744140625,
-0.040924072265625,
-0.0199737548828125,
-0.054168701171875,
-0.01322174072265625,
-0.00608062744140625,
0.018524169921875,
-0.005695343017578125,
-0.0294952392578125,
0.0121002197265625,
0.0034942626953125,
0.040985107421875,
0.018280029296875,
-0.0699462890625,
0.0199737548828125,
0.041229248046875,
0.059173583984375,
-0.0181427001953125,
-0.02581787109375,
0.005451202392578125,
-0.0021953582763671875,
-0.0255279541015625,
0.0675048828125,
-0.0255889892578125,
-0.040283203125,
-0.0191192626953125,
-0.0013914108276367188,
0.0132293701171875,
-0.03717041015625,
0.033477783203125,
-0.0273590087890625,
0.01678466796875,
-0.0278472900390625,
-0.02862548828125,
-0.0245208740234375,
0.0120849609375,
-0.032623291015625,
0.1087646484375,
0.00872802734375,
-0.04022216796875,
0.0235748291015625,
-0.0513916015625,
-0.0142822265625,
-0.01456451416015625,
0.00885772705078125,
-0.040191650390625,
-0.0205230712890625,
0.01105499267578125,
0.02545166015625,
-0.050445556640625,
0.0333251953125,
-0.016632080078125,
-0.0307464599609375,
0.0037517547607421875,
-0.0311431884765625,
0.0631103515625,
0.0204315185546875,
-0.0338134765625,
0.0004019737243652344,
-0.061431884765625,
0.00208282470703125,
0.036376953125,
-0.039306640625,
0.0191802978515625,
0.00545501708984375,
-0.01056671142578125,
0.01517486572265625,
0.037841796875,
-0.02947998046875,
0.0127105712890625,
-0.0244140625,
0.03887939453125,
0.05487060546875,
0.0023555755615234375,
0.01268768310546875,
-0.037353515625,
0.037017822265625,
-0.00391387939453125,
0.028045654296875,
0.0021228790283203125,
-0.056976318359375,
-0.076171875,
-0.0169219970703125,
-0.0005736351013183594,
0.061920166015625,
-0.0218505859375,
0.053558349609375,
-0.002910614013671875,
-0.053680419921875,
-0.033416748046875,
0.0268402099609375,
0.05145263671875,
0.04010009765625,
0.0321044921875,
-0.024200439453125,
-0.04595947265625,
-0.07696533203125,
0.0050201416015625,
-0.033905029296875,
-0.0012769699096679688,
0.0288238525390625,
0.048126220703125,
-0.021820068359375,
0.053680419921875,
-0.04034423828125,
-0.0168304443359375,
-0.022735595703125,
-0.01114654541015625,
0.0077667236328125,
0.02447509765625,
0.04693603515625,
-0.028900146484375,
-0.018157958984375,
-0.0100555419921875,
-0.067138671875,
-0.0093231201171875,
0.00992584228515625,
-0.01274871826171875,
0.02154541015625,
0.023345947265625,
-0.046844482421875,
0.0328369140625,
0.0513916015625,
-0.0141448974609375,
0.0386962890625,
-0.0018177032470703125,
-0.01287841796875,
-0.0775146484375,
0.0038700103759765625,
-0.01514434814453125,
0.0048980712890625,
-0.032928466796875,
-0.00714111328125,
-0.015777587890625,
0.00778961181640625,
-0.045074462890625,
0.045501708984375,
-0.0254058837890625,
-0.0125885009765625,
-0.0092620849609375,
0.004688262939453125,
0.003200531005859375,
0.046295166015625,
-0.006282806396484375,
0.08038330078125,
0.028900146484375,
-0.04254150390625,
0.0163116455078125,
0.03204345703125,
-0.037017822265625,
0.01025390625,
-0.06585693359375,
0.025238037109375,
0.00965118408203125,
0.042999267578125,
-0.07550048828125,
-0.027496337890625,
0.023040771484375,
-0.03533935546875,
0.007587432861328125,
0.01812744140625,
-0.044097900390625,
-0.0310211181640625,
-0.03424072265625,
0.0232391357421875,
0.059783935546875,
-0.034393310546875,
0.01194000244140625,
0.028778076171875,
-0.0002880096435546875,
-0.051666259765625,
-0.062225341796875,
0.007175445556640625,
-0.0243072509765625,
-0.038970947265625,
0.0242767333984375,
-0.01172637939453125,
-0.0158233642578125,
-0.0176544189453125,
0.004940032958984375,
0.0023651123046875,
0.0271759033203125,
0.0261688232421875,
0.029388427734375,
-0.009918212890625,
-0.0015201568603515625,
0.00844573974609375,
-0.0176544189453125,
0.0014257431030273438,
0.010711669921875,
0.046478271484375,
-0.01471710205078125,
-0.01520538330078125,
-0.057769775390625,
-0.0005826950073242188,
0.0224761962890625,
-0.0173797607421875,
0.04827880859375,
0.031219482421875,
-0.01491546630859375,
0.0205230712890625,
-0.05865478515625,
-0.008148193359375,
-0.039337158203125,
0.040618896484375,
-0.0177001953125,
-0.06317138671875,
0.04034423828125,
0.0035228729248046875,
0.02850341796875,
0.057220458984375,
0.046539306640625,
-0.005893707275390625,
0.06121826171875,
0.04351806640625,
-0.00605010986328125,
0.02459716796875,
-0.036163330078125,
-0.005405426025390625,
-0.0723876953125,
-0.04986572265625,
-0.0216522216796875,
-0.03167724609375,
-0.052520751953125,
-0.0290985107421875,
0.0202789306640625,
0.01255035400390625,
-0.050323486328125,
0.0251312255859375,
-0.042205810546875,
0.04193115234375,
0.040863037109375,
0.01139068603515625,
0.0235748291015625,
0.0071258544921875,
0.008544921875,
0.0012865066528320312,
-0.040374755859375,
-0.055999755859375,
0.11138916015625,
0.033905029296875,
0.031707763671875,
0.00827789306640625,
0.04840087890625,
0.0107574462890625,
0.0218505859375,
-0.05126953125,
0.047454833984375,
0.0009279251098632812,
-0.052642822265625,
-0.0147705078125,
-0.0087432861328125,
-0.06756591796875,
0.0116424560546875,
-0.016357421875,
-0.057647705078125,
0.0014104843139648438,
-0.0033416748046875,
-0.028411865234375,
0.0227203369140625,
-0.047607421875,
0.04510498046875,
-0.041748046875,
-0.0231781005859375,
-0.0245513916015625,
-0.06231689453125,
0.050262451171875,
-0.01451873779296875,
0.00878143310546875,
-0.037567138671875,
-0.0255889892578125,
0.068359375,
-0.0291595458984375,
0.07342529296875,
-0.0007376670837402344,
-0.0105743408203125,
0.041229248046875,
-0.0126495361328125,
0.031463623046875,
0.0026683807373046875,
-0.018524169921875,
0.050262451171875,
-0.013519287109375,
-0.0250701904296875,
-0.01018524169921875,
0.038177490234375,
-0.0919189453125,
-0.05743408203125,
-0.037567138671875,
-0.036407470703125,
-0.003215789794921875,
0.005489349365234375,
0.03887939453125,
-0.00670623779296875,
0.00032138824462890625,
0.0106353759765625,
0.03564453125,
-0.041961669921875,
0.036956787109375,
0.04510498046875,
-0.0060272216796875,
-0.04034423828125,
0.052337646484375,
0.0038700103759765625,
0.026824951171875,
0.0191802978515625,
0.0037097930908203125,
-0.0340576171875,
-0.02972412109375,
-0.03631591796875,
0.0253143310546875,
-0.034271240234375,
-0.035247802734375,
-0.043243408203125,
-0.0261993408203125,
-0.0240631103515625,
-0.00417327880859375,
-0.033050537109375,
-0.0343017578125,
-0.0584716796875,
-0.0285797119140625,
0.036712646484375,
0.06292724609375,
-0.0027942657470703125,
0.04595947265625,
-0.0239105224609375,
0.01418304443359375,
0.0295257568359375,
0.014312744140625,
-0.0009465217590332031,
-0.062042236328125,
0.007602691650390625,
0.00803375244140625,
-0.056396484375,
-0.046539306640625,
0.019256591796875,
0.0234527587890625,
0.03424072265625,
0.034698486328125,
-0.00746917724609375,
0.057708740234375,
-0.0274200439453125,
0.08306884765625,
0.0305633544921875,
-0.051055908203125,
0.05047607421875,
-0.016357421875,
0.0035266876220703125,
0.0458984375,
0.019195556640625,
-0.0060882568359375,
-0.0150909423828125,
-0.051300048828125,
-0.051666259765625,
0.061614990234375,
0.0193328857421875,
0.01306915283203125,
0.0037937164306640625,
0.032928466796875,
0.006175994873046875,
0.008575439453125,
-0.06451416015625,
-0.02471923828125,
-0.0219573974609375,
-0.007686614990234375,
-0.015655517578125,
-0.037567138671875,
-0.0011692047119140625,
-0.0246734619140625,
0.04840087890625,
0.00399017333984375,
0.0256500244140625,
-0.0080718994140625,
0.0019931793212890625,
-0.006267547607421875,
0.0027637481689453125,
0.057525634765625,
0.036712646484375,
-0.0187225341796875,
-0.0099029541015625,
0.049407958984375,
-0.0494384765625,
0.0253448486328125,
-0.002094268798828125,
-0.00644683837890625,
-0.0279693603515625,
0.033447265625,
0.06396484375,
0.0188446044921875,
-0.05328369140625,
0.0272216796875,
0.00909423828125,
-0.0274658203125,
-0.02716064453125,
0.0248565673828125,
0.00789642333984375,
0.0230255126953125,
0.022430419921875,
-0.0135345458984375,
0.00574493408203125,
-0.03997802734375,
-0.00902557373046875,
0.0277862548828125,
0.00717926025390625,
-0.032867431640625,
0.07366943359375,
0.0242156982421875,
-0.0199737548828125,
0.039154052734375,
-0.01517486572265625,
-0.030364990234375,
0.066650390625,
0.04852294921875,
0.0467529296875,
-0.0183258056640625,
0.009429931640625,
0.05413818359375,
0.03460693359375,
-0.0164337158203125,
0.018707275390625,
0.0008740425109863281,
-0.03717041015625,
-0.0156097412109375,
-0.052490234375,
-0.03326416015625,
0.030364990234375,
-0.041534423828125,
0.022216796875,
-0.04669189453125,
-0.0200653076171875,
-0.02239990234375,
0.03662109375,
-0.04931640625,
0.0164337158203125,
0.01061248779296875,
0.06939697265625,
-0.056060791015625,
0.058380126953125,
0.038116455078125,
-0.040771484375,
-0.065673828125,
-0.023284912109375,
0.0150604248046875,
-0.09375,
0.041961669921875,
0.0274658203125,
-0.00506591796875,
0.0079345703125,
-0.05572509765625,
-0.092041015625,
0.12646484375,
0.0347900390625,
-0.055206298828125,
0.001735687255859375,
0.0254058837890625,
0.04010009765625,
-0.00937652587890625,
0.033935546875,
0.06341552734375,
0.03656005859375,
0.0091400146484375,
-0.0816650390625,
0.006992340087890625,
-0.027008056640625,
-0.0019817352294921875,
-0.01398468017578125,
-0.099609375,
0.061492919921875,
-0.0290374755859375,
-0.0172882080078125,
0.01528167724609375,
0.0489501953125,
0.051361083984375,
0.036895751953125,
0.026763916015625,
0.062103271484375,
0.0670166015625,
-0.0029087066650390625,
0.08251953125,
-0.025787353515625,
0.01308441162109375,
0.06768798828125,
-0.02239990234375,
0.07220458984375,
0.016845703125,
-0.046234130859375,
0.045074462890625,
0.07757568359375,
-0.0005922317504882812,
0.046875,
0.0022029876708984375,
-0.011016845703125,
-0.0128173828125,
-0.0137176513671875,
-0.04779052734375,
0.037445068359375,
0.020477294921875,
-0.0089569091796875,
-0.0022754669189453125,
-0.0244140625,
0.018280029296875,
-0.024627685546875,
0.00008046627044677734,
0.06396484375,
0.01381683349609375,
-0.046478271484375,
0.066162109375,
0.001590728759765625,
0.0634765625,
-0.045867919921875,
0.007297515869140625,
-0.0400390625,
-0.000008940696716308594,
-0.0262451171875,
-0.056365966796875,
0.003330230712890625,
0.0260467529296875,
-0.0003719329833984375,
-0.00803375244140625,
0.03973388671875,
0.0020275115966796875,
-0.0401611328125,
0.0269317626953125,
0.02093505859375,
0.02642822265625,
0.01593017578125,
-0.05303955078125,
0.01255035400390625,
0.006938934326171875,
-0.038604736328125,
0.0270843505859375,
0.005863189697265625,
-0.0025787353515625,
0.061614990234375,
0.05511474609375,
-0.01361083984375,
0.0088958740234375,
-0.01471710205078125,
0.07232666015625,
-0.03704833984375,
-0.01520538330078125,
-0.058990478515625,
0.03765869140625,
0.003505706787109375,
-0.0550537109375,
0.04034423828125,
0.048614501953125,
0.051055908203125,
0.020782470703125,
0.0482177734375,
0.004215240478515625,
0.0258026123046875,
-0.040863037109375,
0.046295166015625,
-0.058868408203125,
0.0289306640625,
0.005130767822265625,
-0.0740966796875,
-0.006000518798828125,
0.0489501953125,
-0.0174407958984375,
0.0016326904296875,
0.03131103515625,
0.0625,
0.0130157470703125,
-0.01142120361328125,
0.00897979736328125,
0.01424407958984375,
0.025634765625,
0.0660400390625,
0.064208984375,
-0.0465087890625,
0.052459716796875,
-0.0273284912109375,
-0.0176239013671875,
-0.0217437744140625,
-0.055938720703125,
-0.07196044921875,
-0.0206298828125,
-0.01953125,
-0.01171112060546875,
0.005191802978515625,
0.0557861328125,
0.035491943359375,
-0.04339599609375,
-0.0225830078125,
-0.005168914794921875,
-0.006500244140625,
0.0036449432373046875,
-0.0112762451171875,
0.0279693603515625,
-0.010009765625,
-0.040740966796875,
0.0330810546875,
0.004398345947265625,
0.0183563232421875,
-0.0227813720703125,
-0.020965576171875,
-0.01511383056640625,
0.0120391845703125,
0.047760009765625,
0.0221405029296875,
-0.0721435546875,
-0.015960693359375,
0.003116607666015625,
-0.0133514404296875,
0.0096893310546875,
-0.0011987686157226562,
-0.05633544921875,
0.00565338134765625,
0.01033782958984375,
0.027679443359375,
0.0487060546875,
0.00435638427734375,
0.0035877227783203125,
-0.0362548828125,
0.033721923828125,
0.0008902549743652344,
0.01235198974609375,
0.0250701904296875,
-0.0281982421875,
0.06103515625,
0.01371002197265625,
-0.051483154296875,
-0.0711669921875,
0.005619049072265625,
-0.078857421875,
-0.0029621124267578125,
0.1025390625,
-0.000015437602996826172,
-0.01111602783203125,
0.01392364501953125,
-0.01641845703125,
0.0278472900390625,
-0.0303497314453125,
0.0604248046875,
0.043792724609375,
-0.006717681884765625,
-0.005062103271484375,
-0.058837890625,
0.0262603759765625,
0.0304718017578125,
-0.08306884765625,
-0.0205078125,
0.032684326171875,
0.0377197265625,
-0.00678253173828125,
0.051910400390625,
0.0004000663757324219,
0.0177154541015625,
0.0041961669921875,
0.00750732421875,
-0.0175628662109375,
-0.01274871826171875,
-0.00734710693359375,
-0.01971435546875,
-0.003910064697265625,
-0.01690673828125
]
] |
totally-not-an-llm/PuddleJumper-13b-V2 | 2023-09-25T01:16:26.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:totally-not-an-llm/EverythingLM-data-V3",
"dataset:Open-Orca/OpenOrca",
"dataset:garage-bAInd/Open-Platypus",
"license:other",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | totally-not-an-llm | null | null | totally-not-an-llm/PuddleJumper-13b-V2 | 1 | 7,491 | transformers | 2023-09-21T03:42:01 | ---
license: other
datasets:
- totally-not-an-llm/EverythingLM-data-V3
- Open-Orca/OpenOrca
- garage-bAInd/Open-Platypus
---
Merge of EverythingLM-V3-13b QLoRa and OpenOrca-Platypus2-13B.
### Prompt format:
```
USER: <prompt>
ASSISTANT:
```
### Quants:
https://huggingface.co/TheBloke/PuddleJumper-13B-V2-GGUF
https://huggingface.co/TheBloke/PuddleJumper-13B-V2-AWQ
https://huggingface.co/TheBloke/PuddleJumper-13B-V2-GPTQ | 427 | [
[
-0.022613525390625,
-0.028564453125,
0.03839111328125,
0.053863525390625,
-0.038818359375,
-0.00737762451171875,
0.01430511474609375,
-0.0214691162109375,
0.03466796875,
0.04046630859375,
-0.035369873046875,
-0.02117919921875,
-0.0338134765625,
-0.00959014892578125,
-0.0019245147705078125,
0.07916259765625,
-0.01192474365234375,
-0.006404876708984375,
0.0094451904296875,
-0.0272979736328125,
-0.03533935546875,
-0.0229949951171875,
-0.06121826171875,
-0.0029754638671875,
0.049896240234375,
0.0285797119140625,
0.07318115234375,
0.0231170654296875,
0.004547119140625,
0.0231475830078125,
-0.006778717041015625,
0.0201416015625,
-0.0308990478515625,
0.023590087890625,
-0.022125244140625,
-0.0021572113037109375,
-0.04669189453125,
0.0050048828125,
0.05712890625,
0.018157958984375,
-0.0189208984375,
0.01235198974609375,
-0.00392913818359375,
0.03363037109375,
-0.0535888671875,
0.031158447265625,
-0.001087188720703125,
0.0076751708984375,
-0.005889892578125,
-0.0188140869140625,
-0.00688934326171875,
-0.0711669921875,
-0.0180511474609375,
-0.0794677734375,
-0.0016803741455078125,
0.025146484375,
0.0841064453125,
-0.00995635986328125,
-0.043731689453125,
-0.0142669677734375,
-0.03375244140625,
0.037567138671875,
-0.02978515625,
0.0074920654296875,
0.0014352798461914062,
0.0267333984375,
-0.0276641845703125,
-0.07427978515625,
-0.04388427734375,
0.007312774658203125,
-0.01102447509765625,
0.03399658203125,
-0.0462646484375,
-0.00872039794921875,
-0.007427215576171875,
0.0316162109375,
-0.054595947265625,
-0.00626373291015625,
-0.0628662109375,
-0.025054931640625,
0.04034423828125,
0.0280914306640625,
0.0272216796875,
0.0011434555053710938,
-0.0374755859375,
-0.03778076171875,
-0.032135009765625,
-0.00922393798828125,
0.01047515869140625,
0.01084136962890625,
-0.01348876953125,
0.05804443359375,
0.00270843505859375,
0.049072265625,
-0.00687408447265625,
-0.0053558349609375,
0.0173187255859375,
-0.03399658203125,
-0.029052734375,
0.002338409423828125,
0.0654296875,
0.041778564453125,
-0.0019969940185546875,
0.006565093994140625,
-0.00409698486328125,
-0.015716552734375,
-0.0186309814453125,
-0.040283203125,
-0.04522705078125,
0.03887939453125,
-0.0277099609375,
-0.0141448974609375,
0.0556640625,
-0.0428466796875,
-0.02069091796875,
-0.01462554931640625,
0.039215087890625,
-0.0310211181640625,
-0.0374755859375,
0.0214691162109375,
-0.03839111328125,
0.03253173828125,
0.052581787109375,
-0.01580810546875,
0.0181427001953125,
0.0489501953125,
0.041961669921875,
0.0251312255859375,
-0.033172607421875,
-0.036651611328125,
0.019744873046875,
-0.0262451171875,
0.04803466796875,
0.01438140869140625,
-0.039764404296875,
-0.0036220550537109375,
0.021270751953125,
-0.005340576171875,
-0.037384033203125,
0.04852294921875,
-0.033538818359375,
0.057281494140625,
-0.0218963623046875,
-0.01434326171875,
-0.0230712890625,
-0.0090179443359375,
-0.0723876953125,
0.06341552734375,
0.05389404296875,
-0.05670166015625,
-0.01326751708984375,
-0.052947998046875,
0.0024585723876953125,
0.0301513671875,
0.0255126953125,
-0.0172271728515625,
0.0003628730773925781,
-0.01593017578125,
0.013916015625,
-0.0162353515625,
-0.005962371826171875,
-0.0164947509765625,
0.00023925304412841797,
0.01302337646484375,
-0.0011882781982421875,
0.09246826171875,
0.031158447265625,
0.01140594482421875,
0.031585693359375,
-0.036773681640625,
0.0038204193115234375,
0.0411376953125,
-0.00702667236328125,
-0.0141143798828125,
-0.06402587890625,
0.042724609375,
0.0249176025390625,
0.0601806640625,
-0.0158233642578125,
0.03619384765625,
0.0035266876220703125,
0.029296875,
0.040863037109375,
0.01203155517578125,
0.0435791015625,
-0.042572021484375,
0.052093505859375,
0.00977325439453125,
0.03369140625,
0.01287078857421875,
-0.06298828125,
-0.045989990234375,
-0.035888671875,
0.0096282958984375,
0.05084228515625,
-0.04510498046875,
0.053009033203125,
0.0055999755859375,
-0.0528564453125,
-0.032379150390625,
-0.003993988037109375,
0.020263671875,
0.01983642578125,
0.024932861328125,
-0.01236724853515625,
-0.042388916015625,
-0.042236328125,
0.010406494140625,
-0.04541015625,
-0.0214691162109375,
0.0113372802734375,
0.01340484619140625,
-0.01270294189453125,
0.05657958984375,
-0.0601806640625,
-0.032440185546875,
-0.00269317626953125,
-0.0007748603820800781,
0.00876617431640625,
0.04522705078125,
0.06781005859375,
-0.03533935546875,
-0.035919189453125,
-0.003902435302734375,
-0.03155517578125,
-0.021392822265625,
0.01316070556640625,
-0.041229248046875,
0.006988525390625,
0.0004055500030517578,
-0.08587646484375,
0.043182373046875,
0.03192138671875,
-0.0716552734375,
0.045318603515625,
-0.0258331298828125,
0.0217437744140625,
-0.072998046875,
0.0094146728515625,
-0.0149383544921875,
-0.0175018310546875,
-0.037200927734375,
0.028900146484375,
0.0203094482421875,
0.01534271240234375,
-0.0645751953125,
0.06787109375,
-0.044952392578125,
-0.0016222000122070312,
0.007320404052734375,
0.02685546875,
0.0265350341796875,
0.005451202392578125,
-0.03656005859375,
0.03021240234375,
0.056640625,
-0.0227508544921875,
0.05853271484375,
0.0545654296875,
0.028411865234375,
0.01023101806640625,
-0.04443359375,
-0.01678466796875,
-0.005619049072265625,
0.03826904296875,
-0.0848388671875,
-0.0244903564453125,
0.0565185546875,
-0.0217742919921875,
-0.0054168701171875,
0.01204681396484375,
-0.0499267578125,
-0.0289306640625,
-0.056732177734375,
0.0501708984375,
0.06524658203125,
-0.046539306640625,
0.0584716796875,
0.0004451274871826172,
-0.025390625,
-0.00594329833984375,
-0.0401611328125,
-0.00838470458984375,
-0.021331787109375,
-0.044830322265625,
0.028045654296875,
-0.008819580078125,
-0.0169830322265625,
0.003475189208984375,
0.005565643310546875,
-0.012237548828125,
-0.0296783447265625,
0.0209808349609375,
0.0535888671875,
-0.032318115234375,
-0.035888671875,
-0.00015151500701904297,
-0.000009059906005859375,
-0.002933502197265625,
-0.0141448974609375,
0.034759521484375,
-0.007427215576171875,
-0.01507568359375,
-0.020904541015625,
0.0190277099609375,
0.06201171875,
-0.01065826416015625,
0.057159423828125,
0.05401611328125,
-0.01526641845703125,
0.026885986328125,
-0.0469970703125,
0.00408935546875,
-0.0295562744140625,
-0.009918212890625,
-0.011871337890625,
-0.07891845703125,
0.06341552734375,
0.0255279541015625,
-0.006000518798828125,
0.0518798828125,
0.0227508544921875,
0.006732940673828125,
0.054718017578125,
0.032318115234375,
0.0023651123046875,
0.032379150390625,
-0.01235198974609375,
0.0283203125,
-0.056182861328125,
-0.03369140625,
-0.037139892578125,
0.00283050537109375,
-0.03961181640625,
-0.0238800048828125,
0.0034999847412109375,
0.05181884765625,
-0.0282135009765625,
0.046844482421875,
-0.03564453125,
0.0357666015625,
0.04962158203125,
0.0136260986328125,
0.0193634033203125,
-0.0235137939453125,
0.021148681640625,
0.0207366943359375,
-0.046600341796875,
-0.0164642333984375,
0.050262451171875,
0.0279388427734375,
0.015869140625,
0.041015625,
0.03924560546875,
-0.015899658203125,
0.00897979736328125,
-0.034637451171875,
0.040985107421875,
0.005786895751953125,
-0.02777099609375,
-0.0200653076171875,
-0.00775146484375,
-0.09112548828125,
0.0208740234375,
-0.0006175041198730469,
-0.054901123046875,
0.01091766357421875,
-0.0006890296936035156,
-0.0408935546875,
0.0098724365234375,
-0.04400634765625,
0.058868408203125,
0.0180816650390625,
-0.008514404296875,
-0.00707244873046875,
-0.0465087890625,
0.06536865234375,
-0.01027679443359375,
-0.0014829635620117188,
-0.0003523826599121094,
-0.0286865234375,
0.04364013671875,
-0.06060791015625,
0.00615692138671875,
-0.0182647705078125,
-0.0218963623046875,
0.035888671875,
0.017242431640625,
0.0158538818359375,
0.0241241455078125,
0.03009033203125,
-0.0010509490966796875,
0.0128631591796875,
-0.047454833984375,
-0.03289794921875,
0.054901123046875,
-0.06256103515625,
-0.01213836669921875,
-0.04949951171875,
-0.0171661376953125,
0.0215606689453125,
-0.00797271728515625,
0.0202484130859375,
0.028472900390625,
0.000396728515625,
-0.0027370452880859375,
0.033599853515625,
-0.007610321044921875,
0.04156494140625,
0.0399169921875,
-0.058990478515625,
-0.06396484375,
0.05157470703125,
-0.017547607421875,
-0.002277374267578125,
0.00983428955078125,
0.01148223876953125,
-0.055328369140625,
-0.0161590576171875,
-0.01116943359375,
0.03436279296875,
-0.045013427734375,
-0.051055908203125,
-0.03973388671875,
-0.042205810546875,
-0.034942626953125,
-0.007236480712890625,
0.005168914794921875,
-0.050628662109375,
-0.043212890625,
-0.0009794235229492188,
0.06927490234375,
0.059326171875,
-0.04022216796875,
0.06402587890625,
-0.0841064453125,
0.0247039794921875,
0.03717041015625,
-0.005825042724609375,
-0.001552581787109375,
-0.045928955078125,
0.0084228515625,
-0.005390167236328125,
-0.037078857421875,
-0.0765380859375,
0.047821044921875,
0.01053619384765625,
0.02777099609375,
0.034942626953125,
0.005718231201171875,
0.06201171875,
-0.00968170166015625,
0.07098388671875,
0.031707763671875,
-0.060089111328125,
0.03326416015625,
-0.03338623046875,
0.0106353759765625,
0.0139007568359375,
0.0225677490234375,
-0.011749267578125,
-0.038909912109375,
-0.0869140625,
-0.07525634765625,
0.048187255859375,
0.04742431640625,
-0.0272216796875,
0.00455474853515625,
-0.003894805908203125,
0.0022640228271484375,
0.015594482421875,
-0.03564453125,
-0.050567626953125,
-0.02447509765625,
-0.0118560791015625,
0.0016622543334960938,
-0.0242767333984375,
-0.02545166015625,
-0.03619384765625,
0.047271728515625,
0.0082855224609375,
0.00501251220703125,
0.00167083740234375,
0.00875091552734375,
-0.0228118896484375,
0.0019550323486328125,
0.0615234375,
0.045806884765625,
-0.037811279296875,
-0.0086669921875,
-0.0128173828125,
-0.037017822265625,
0.00514984130859375,
0.01280975341796875,
0.029510498046875,
0.0154876708984375,
0.02191162109375,
0.05230712890625,
0.014801025390625,
-0.0207366943359375,
0.04974365234375,
-0.0236663818359375,
-0.022674560546875,
-0.021636962890625,
0.0022640228271484375,
0.0128631591796875,
0.0291748046875,
0.026214599609375,
0.01120758056640625,
0.0050201416015625,
-0.0399169921875,
0.00992584228515625,
0.01523590087890625,
-0.00304412841796875,
-0.0152740478515625,
0.045562744140625,
-0.004756927490234375,
-0.014892578125,
0.044036865234375,
-0.034759521484375,
-0.041412353515625,
0.039154052734375,
0.048858642578125,
0.07659912109375,
-0.0208740234375,
0.0140228271484375,
0.0248565673828125,
-0.000629425048828125,
-0.01239776611328125,
0.055145263671875,
0.01026153564453125,
-0.037689208984375,
-0.0181884765625,
-0.01849365234375,
-0.0220184326171875,
0.01235198974609375,
-0.05767822265625,
0.031402587890625,
-0.042694091796875,
-0.0038700103759765625,
-0.0159912109375,
0.004695892333984375,
-0.038787841796875,
0.0036640167236328125,
-0.006099700927734375,
0.06475830078125,
-0.08612060546875,
0.055267333984375,
0.06793212890625,
-0.047332763671875,
-0.07061767578125,
-0.02264404296875,
-0.004138946533203125,
-0.055694580078125,
0.028900146484375,
-0.0258941650390625,
0.03094482421875,
-0.001186370849609375,
-0.04931640625,
-0.059722900390625,
0.10394287109375,
0.017578125,
-0.0231781005859375,
0.02667236328125,
-0.0234832763671875,
0.0183868408203125,
-0.0230255126953125,
0.0516357421875,
0.04156494140625,
0.05987548828125,
0.039581298828125,
-0.08770751953125,
0.0205078125,
-0.01366424560546875,
0.0034942626953125,
0.0201416015625,
-0.06494140625,
0.08685302734375,
0.00384521484375,
-0.0218963623046875,
0.0364990234375,
0.05224609375,
0.0660400390625,
0.01181793212890625,
0.0230255126953125,
0.044952392578125,
0.0384521484375,
-0.039825439453125,
0.0438232421875,
-0.01025390625,
0.03302001953125,
0.0723876953125,
-0.0247802734375,
0.0447998046875,
0.0308685302734375,
-0.0338134765625,
0.031768798828125,
0.06695556640625,
-0.00452423095703125,
0.0423583984375,
-0.004848480224609375,
-0.0175323486328125,
0.01154327392578125,
-0.0004355907440185547,
-0.05804443359375,
-0.01270294189453125,
0.0117645263671875,
0.01073455810546875,
-0.032806396484375,
-0.004215240478515625,
0.0172576904296875,
-0.022369384765625,
-0.02423095703125,
0.059173583984375,
-0.0010747909545898438,
-0.039581298828125,
0.03570556640625,
0.0028553009033203125,
0.045928955078125,
-0.041229248046875,
-0.004730224609375,
-0.0250396728515625,
-0.0037384033203125,
-0.0233612060546875,
-0.08154296875,
0.00418853759765625,
-0.0400390625,
0.00481414794921875,
-0.01192474365234375,
0.04913330078125,
-0.043731689453125,
-0.00701904296875,
0.01526641845703125,
0.055084228515625,
0.0211334228515625,
-0.0165863037109375,
-0.052398681640625,
0.0195465087890625,
0.0196685791015625,
-0.0019140243530273438,
0.0232086181640625,
0.054168701171875,
0.0077362060546875,
0.037017822265625,
0.046600341796875,
0.01371002197265625,
0.01715087890625,
0.01103973388671875,
0.072265625,
-0.048187255859375,
-0.0335693359375,
-0.05804443359375,
0.045745849609375,
0.002498626708984375,
-0.039306640625,
0.04010009765625,
0.052947998046875,
0.0404052734375,
-0.038177490234375,
0.045928955078125,
-0.01265716552734375,
0.0068817138671875,
-0.0281829833984375,
0.03936767578125,
-0.03350830078125,
-0.02520751953125,
0.00335693359375,
-0.0953369140625,
-0.0146942138671875,
0.036956787109375,
0.034881591796875,
-0.006046295166015625,
0.07330322265625,
0.04339599609375,
-0.0128173828125,
0.0026988983154296875,
0.01035308837890625,
0.01235198974609375,
0.01270294189453125,
0.048736572265625,
0.0679931640625,
-0.060943603515625,
0.01654052734375,
-0.0266265869140625,
-0.02099609375,
-0.035003662109375,
-0.06573486328125,
-0.04913330078125,
-0.036712646484375,
-0.036285400390625,
-0.049041748046875,
-0.01593017578125,
0.0814208984375,
0.0160980224609375,
-0.0777587890625,
-0.019134521484375,
0.0024776458740234375,
-0.0010786056518554688,
0.00788116455078125,
-0.0221710205078125,
0.0330810546875,
0.0134735107421875,
-0.045745849609375,
0.0019207000732421875,
0.01503753662109375,
0.0614013671875,
-0.0216064453125,
-0.0259857177734375,
0.01488494873046875,
-0.0005040168762207031,
0.041229248046875,
0.043975830078125,
-0.082275390625,
0.006046295166015625,
-0.0248260498046875,
-0.01262664794921875,
-0.00011551380157470703,
0.02685546875,
-0.04486083984375,
-0.023284912109375,
0.07086181640625,
-0.0136566162109375,
0.0148773193359375,
0.020599365234375,
0.039886474609375,
-0.041015625,
0.031494140625,
0.00389862060546875,
0.0382080078125,
0.01038360595703125,
-0.010498046875,
0.03759765625,
0.0087890625,
-0.039276123046875,
-0.05780029296875,
-0.0216064453125,
-0.108154296875,
-0.0119476318359375,
0.07391357421875,
-0.001270294189453125,
-0.0283966064453125,
0.003993988037109375,
-0.07537841796875,
0.032196044921875,
-0.05096435546875,
0.0183868408203125,
0.039093017578125,
-0.0100555419921875,
-0.00833892822265625,
-0.01263427734375,
0.0311126708984375,
0.01422882080078125,
-0.0517578125,
-0.016998291015625,
0.0335693359375,
0.0090484619140625,
0.0239715576171875,
0.03265380859375,
-0.006488800048828125,
0.0228118896484375,
-0.01303863525390625,
0.021697998046875,
-0.01242828369140625,
-0.0117645263671875,
-0.039276123046875,
0.01371002197265625,
0.01375579833984375,
-0.046661376953125
]
] |
speechlessai/speechless-codellama-airoboros-orca-platypus-13b | 2023-09-22T05:33:29.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-2",
"en",
"dataset:jondurbin/airoboros-2.2",
"dataset:Open-Orca/OpenOrca",
"dataset:garage-bAInd/Open-Platypus",
"arxiv:2308.12950",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | speechlessai | null | null | speechlessai/speechless-codellama-airoboros-orca-platypus-13b | 0 | 7,488 | transformers | 2023-09-19T09:29:53 | ---
language:
- en
library_name: transformers
pipeline_tag: text-generation
datasets:
- jondurbin/airoboros-2.2
- Open-Orca/OpenOrca
- garage-bAInd/Open-Platypus
tags:
- llama-2
license: llama2
---
<p><h1> speechless-codellama-airoboros-orca-platypus-13b </h1></p>
Use the following dataset to fine-tune codellama/CodeLlama-13B in order to improve the model's reasoning and planning abilities.
- jondurbin/airoboros-2.2: Filter categories related to coding, reasoning and planning.
- Open-Orca/OpenOrca: Filter the 'cot' category in 1M GPT4 dataset.
- garage-bAInd/Open-Platypus: 100%
| Metric | Value |
| --- | --- |
| humaneval-python | 49.39 |
[Big Code Models Leaderboard](https://huggingface.co/spaces/bigcode/bigcode-models-leaderboard)
CodeLlama-34B-Python: 53.29
CodeLlama-34B-Instruct: 50.79
CodeLlama-13B-Instruct: 50.6
CodeLlama-34B: 45.11
CodeLlama-13B-Python: 42.89
CodeLlama-13B: 35.07
[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
| Metric | Value |
| --- | --- |
| ARC | 44.88 |
| HellaSwag | 67.7 |
| MMLU | 43.16 |
| TruthfulQA | 40.88 |
| Average | 49.15 |
# **Code Llama**
Code Llama is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 34 billion parameters. This is the repository for the base 13B version in the Hugging Face Transformers format. This model is designed for general code synthesis and understanding. Links to other models can be found in the index at the bottom.
| | Base Model | Python | Instruct |
| --- | ----------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
| 7B | [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) | [codellama/CodeLlama-7b-Python-hf](https://huggingface.co/codellama/CodeLlama-7b-Python-hf) | [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) |
| 13B | [codellama/CodeLlama-13b-hf](https://huggingface.co/codellama/CodeLlama-13b-hf) | [codellama/CodeLlama-13b-Python-hf](https://huggingface.co/codellama/CodeLlama-13b-Python-hf) | [codellama/CodeLlama-13b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-13b-Instruct-hf) |
| 34B | [codellama/CodeLlama-34b-hf](https://huggingface.co/codellama/CodeLlama-34b-hf) | [codellama/CodeLlama-34b-Python-hf](https://huggingface.co/codellama/CodeLlama-34b-Python-hf) | [codellama/CodeLlama-34b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-34b-Instruct-hf) |
## Model Use
To use this model, please make sure to install transformers from `main` until the next version is released:
```bash
pip install git+https://github.com/huggingface/transformers.git@main accelerate
```
Model capabilities:
- [x] Code completion.
- [x] Infilling.
- [ ] Instructions / chat.
- [ ] Python specialist.
```python
from transformers import AutoTokenizer
import transformers
import torch
model = "codellama/CodeLlama-13b-hf"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
sequences = pipeline(
'import socket\n\ndef ping_exponential_backoff(host: str):',
do_sample=True,
top_k=10,
temperature=0.1,
top_p=0.95,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
max_length=200,
)
for seq in sequences:
print(f"Result: {seq['generated_text']}")
```
## Model Details
*Note: Use of this model is governed by the Meta license. Meta developed and publicly released the Code Llama family of large language models (LLMs).
**Model Developers** Meta
**Variations** Code Llama comes in three model sizes, and three variants:
* Code Llama: base models designed for general code synthesis and understanding
* Code Llama - Python: designed specifically for Python
* Code Llama - Instruct: for instruction following and safer deployment
All variants are available in sizes of 7B, 13B and 34B parameters.
**This repository contains the base version of the 13B parameters model.**
**Input** Models input text only.
**Output** Models generate text only.
**Model Architecture** Code Llama is an auto-regressive language model that uses an optimized transformer architecture.
**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.
**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** 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/)" or its [arXiv page](https://arxiv.org/abs/2308.12950).
## 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).
## 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).
| 7,825 | [
[
-0.030792236328125,
-0.050506591796875,
0.0220947265625,
0.03778076171875,
-0.01476287841796875,
0.006114959716796875,
-0.01459503173828125,
-0.043304443359375,
0.0183563232421875,
0.034332275390625,
-0.0313720703125,
-0.046875,
-0.04132080078125,
0.0161590576171875,
-0.027557373046875,
0.0823974609375,
-0.004787445068359375,
-0.0261383056640625,
-0.01221466064453125,
0.0011148452758789062,
-0.02581787109375,
-0.043975830078125,
-0.0222015380859375,
-0.033660888671875,
0.0186920166015625,
0.0236663818359375,
0.05389404296875,
0.046142578125,
0.039886474609375,
0.0265655517578125,
-0.0247344970703125,
0.00771331787109375,
-0.0247650146484375,
-0.021820068359375,
0.010345458984375,
-0.03631591796875,
-0.057586669921875,
-0.00469970703125,
0.0281219482421875,
0.02459716796875,
-0.0214080810546875,
0.02789306640625,
-0.00574493408203125,
0.040374755859375,
-0.023468017578125,
0.021697998046875,
-0.04766845703125,
-0.00695037841796875,
0.0045166015625,
-0.01482391357421875,
-0.015838623046875,
-0.036407470703125,
-0.00879669189453125,
-0.035736083984375,
-0.0013093948364257812,
-0.00122833251953125,
0.09222412109375,
0.0328369140625,
-0.016845703125,
-0.0195159912109375,
-0.028717041015625,
0.05810546875,
-0.0718994140625,
0.006717681884765625,
0.02655029296875,
-0.00811767578125,
-0.01537322998046875,
-0.0574951171875,
-0.0518798828125,
-0.02508544921875,
-0.01198577880859375,
0.005222320556640625,
-0.0277557373046875,
0.002422332763671875,
0.031280517578125,
0.03369140625,
-0.040130615234375,
0.00910186767578125,
-0.0333251953125,
-0.0192718505859375,
0.06854248046875,
0.0144500732421875,
0.029327392578125,
-0.02459716796875,
-0.0214080810546875,
-0.00860595703125,
-0.057098388671875,
0.01053619384765625,
0.032196044921875,
-0.0023670196533203125,
-0.04803466796875,
0.047698974609375,
-0.0184173583984375,
0.049560546875,
0.00614166259765625,
-0.034881591796875,
0.044036865234375,
-0.0286712646484375,
-0.024139404296875,
-0.01396942138671875,
0.065673828125,
0.036590576171875,
0.0220794677734375,
0.005970001220703125,
-0.0177001953125,
0.02191162109375,
0.004779815673828125,
-0.0654296875,
-0.0166168212890625,
0.0271148681640625,
-0.044830322265625,
-0.044891357421875,
-0.01477813720703125,
-0.060821533203125,
-0.0018873214721679688,
-0.0108642578125,
0.0196990966796875,
-0.0240325927734375,
-0.03375244140625,
0.023040771484375,
0.001216888427734375,
0.03179931640625,
0.00934600830078125,
-0.0672607421875,
0.003475189208984375,
0.03631591796875,
0.059814453125,
0.0067596435546875,
-0.03436279296875,
-0.0016603469848632812,
-0.00765228271484375,
-0.022796630859375,
0.048248291015625,
-0.0303192138671875,
-0.027740478515625,
-0.01438140869140625,
0.0086822509765625,
-0.00920867919921875,
-0.035552978515625,
0.0172271728515625,
-0.0240936279296875,
0.00554656982421875,
0.005420684814453125,
-0.0267791748046875,
-0.027923583984375,
0.00322723388671875,
-0.0389404296875,
0.0831298828125,
0.017578125,
-0.0556640625,
0.0007119178771972656,
-0.0482177734375,
-0.026611328125,
-0.0205078125,
-0.0007405281066894531,
-0.05426025390625,
-0.0081024169921875,
0.02410888671875,
0.039886474609375,
-0.0240020751953125,
0.0293731689453125,
-0.0137786865234375,
-0.0267333984375,
0.013824462890625,
-0.0085601806640625,
0.07806396484375,
0.025421142578125,
-0.0386962890625,
0.0186767578125,
-0.0614013671875,
0.002368927001953125,
0.03424072265625,
-0.034576416015625,
0.0108795166015625,
-0.00934600830078125,
0.0000603795051574707,
0.00511932373046875,
0.0333251953125,
-0.0231475830078125,
0.03350830078125,
-0.034515380859375,
0.062255859375,
0.05206298828125,
0.0020503997802734375,
0.028167724609375,
-0.0435791015625,
0.050384521484375,
-0.0038700103759765625,
0.0182952880859375,
-0.02093505859375,
-0.055999755859375,
-0.07415771484375,
-0.0251617431640625,
0.0096588134765625,
0.049346923828125,
-0.04351806640625,
0.055419921875,
-0.0034942626953125,
-0.058258056640625,
-0.037872314453125,
0.0138092041015625,
0.030792236328125,
0.02740478515625,
0.030609130859375,
-0.01473236083984375,
-0.05975341796875,
-0.0567626953125,
0.006969451904296875,
-0.03106689453125,
0.01343536376953125,
0.01548004150390625,
0.0631103515625,
-0.039276123046875,
0.06280517578125,
-0.040435791015625,
-0.007343292236328125,
-0.02728271484375,
-0.021820068359375,
0.045135498046875,
0.049224853515625,
0.0517578125,
-0.049285888671875,
-0.0222930908203125,
0.0011768341064453125,
-0.0640869140625,
-0.00858306884765625,
-0.01605224609375,
-0.00704193115234375,
0.0278778076171875,
0.025360107421875,
-0.049957275390625,
0.041290283203125,
0.053802490234375,
-0.024261474609375,
0.049407958984375,
-0.01340484619140625,
-0.006183624267578125,
-0.076904296875,
0.02093505859375,
-0.00641632080078125,
-0.0015954971313476562,
-0.035308837890625,
0.0266876220703125,
0.00867462158203125,
0.0067138671875,
-0.034454345703125,
0.032470703125,
-0.032257080078125,
0.00316619873046875,
-0.00799560546875,
-0.00994873046875,
0.0030574798583984375,
0.05596923828125,
0.003452301025390625,
0.06634521484375,
0.047210693359375,
-0.04730224609375,
0.0294342041015625,
0.024871826171875,
-0.025970458984375,
0.015777587890625,
-0.0693359375,
0.01556396484375,
0.006969451904296875,
0.0285491943359375,
-0.06781005859375,
-0.0170440673828125,
0.025421142578125,
-0.04229736328125,
0.01267242431640625,
-0.0028247833251953125,
-0.037322998046875,
-0.04345703125,
-0.0196075439453125,
0.027557373046875,
0.0626220703125,
-0.04730224609375,
0.031402587890625,
0.028289794921875,
0.009368896484375,
-0.045379638671875,
-0.04876708984375,
0.0044708251953125,
-0.034698486328125,
-0.059967041015625,
0.0333251953125,
-0.0213165283203125,
-0.0059967041015625,
-0.01273345947265625,
0.0022106170654296875,
0.0008869171142578125,
0.0195465087890625,
0.031707763671875,
0.037139892578125,
-0.012481689453125,
-0.01094818115234375,
-0.0076446533203125,
-0.013519287109375,
0.00409698486328125,
-0.0014581680297851562,
0.055755615234375,
-0.0296478271484375,
-0.0119171142578125,
-0.043212890625,
0.008331298828125,
0.04327392578125,
-0.0244598388671875,
0.050506591796875,
0.03570556640625,
-0.0237579345703125,
-0.0034770965576171875,
-0.04827880859375,
-0.002017974853515625,
-0.041656494140625,
0.022369384765625,
-0.022613525390625,
-0.060516357421875,
0.05023193359375,
0.0109100341796875,
0.01103973388671875,
0.0390625,
0.0595703125,
0.006458282470703125,
0.06005859375,
0.061309814453125,
-0.0270843505859375,
0.03131103515625,
-0.04974365234375,
0.0091552734375,
-0.0626220703125,
-0.031768798828125,
-0.044281005859375,
-0.00604248046875,
-0.04852294921875,
-0.0301971435546875,
0.0260009765625,
0.01306915283203125,
-0.0345458984375,
0.055755615234375,
-0.06591796875,
0.0295562744140625,
0.03424072265625,
0.01210784912109375,
0.023040771484375,
0.004718780517578125,
-0.007236480712890625,
0.01812744140625,
-0.04412841796875,
-0.0498046875,
0.08892822265625,
0.035491943359375,
0.0657958984375,
-0.004306793212890625,
0.060455322265625,
-0.002704620361328125,
0.021636962890625,
-0.043670654296875,
0.042816162109375,
0.016754150390625,
-0.0377197265625,
-0.004425048828125,
-0.0268707275390625,
-0.06982421875,
0.01482391357421875,
0.00186920166015625,
-0.0631103515625,
0.00786590576171875,
0.007781982421875,
-0.02362060546875,
0.026885986328125,
-0.055389404296875,
0.055389404296875,
-0.008758544921875,
-0.00792694091796875,
-0.00930023193359375,
-0.041778564453125,
0.036590576171875,
-0.000027418136596679688,
0.01355743408203125,
-0.00788116455078125,
-0.0089874267578125,
0.0517578125,
-0.043701171875,
0.07470703125,
0.0048370361328125,
-0.0285797119140625,
0.04327392578125,
-0.005390167236328125,
0.03265380859375,
0.0035686492919921875,
-0.021270751953125,
0.0435791015625,
-0.0013790130615234375,
-0.0217132568359375,
-0.01092529296875,
0.05206298828125,
-0.07769775390625,
-0.054656982421875,
-0.031158447265625,
-0.0291290283203125,
0.016021728515625,
0.00762939453125,
0.034210205078125,
0.0149078369140625,
0.01552581787109375,
0.0120697021484375,
0.035858154296875,
-0.046417236328125,
0.049468994140625,
0.0288848876953125,
-0.0183563232421875,
-0.0380859375,
0.06524658203125,
-0.005615234375,
0.019378662109375,
0.0198516845703125,
0.00466156005859375,
-0.01514434814453125,
-0.034637451171875,
-0.032440185546875,
0.032806396484375,
-0.043853759765625,
-0.0452880859375,
-0.051849365234375,
-0.03436279296875,
-0.034942626953125,
-0.0177764892578125,
-0.017913818359375,
-0.0238189697265625,
-0.046478271484375,
-0.01111602783203125,
0.055572509765625,
0.04833984375,
-0.005611419677734375,
0.0281982421875,
-0.04547119140625,
0.030517578125,
0.007137298583984375,
0.0291900634765625,
0.006160736083984375,
-0.044769287109375,
-0.011138916015625,
0.0005602836608886719,
-0.0445556640625,
-0.06494140625,
0.04522705078125,
0.00922393798828125,
0.04132080078125,
0.004425048828125,
-0.0033512115478515625,
0.0552978515625,
-0.0276947021484375,
0.0626220703125,
0.0225982666015625,
-0.08734130859375,
0.05291748046875,
-0.0209808349609375,
0.01070404052734375,
0.00753021240234375,
0.019927978515625,
-0.032928466796875,
-0.0277099609375,
-0.05438232421875,
-0.062164306640625,
0.05499267578125,
0.0260772705078125,
0.01092529296875,
0.005985260009765625,
0.03228759765625,
-0.0093231201171875,
0.0181427001953125,
-0.07904052734375,
-0.023468017578125,
-0.0260772705078125,
-0.0148162841796875,
-0.0050048828125,
-0.0108184814453125,
0.00045609474182128906,
-0.0287017822265625,
0.044586181640625,
-0.00933074951171875,
0.041748046875,
0.0195159912109375,
-0.00980377197265625,
-0.01898193359375,
0.002758026123046875,
0.04931640625,
0.049224853515625,
-0.005588531494140625,
-0.010894775390625,
0.024658203125,
-0.0386962890625,
0.0167388916015625,
-0.006031036376953125,
-0.0085296630859375,
-0.015533447265625,
0.0408935546875,
0.050445556640625,
0.002140045166015625,
-0.05047607421875,
0.03729248046875,
0.0081787109375,
-0.0196533203125,
-0.03302001953125,
0.02374267578125,
0.02490234375,
0.020172119140625,
0.016998291015625,
0.0023326873779296875,
-0.01082611083984375,
-0.033050537109375,
0.006374359130859375,
0.021942138671875,
0.01123809814453125,
-0.02447509765625,
0.065673828125,
0.00933837890625,
-0.020111083984375,
0.034912109375,
-0.00024139881134033203,
-0.05157470703125,
0.09161376953125,
0.0469970703125,
0.05450439453125,
-0.019683837890625,
0.01065826416015625,
0.044342041015625,
0.04193115234375,
-0.0014104843139648438,
0.03204345703125,
0.00536346435546875,
-0.0421142578125,
-0.025421142578125,
-0.059295654296875,
-0.0167694091796875,
0.00792694091796875,
-0.035491943359375,
0.0299072265625,
-0.047210693359375,
-0.00638580322265625,
-0.017333984375,
0.014495849609375,
-0.052581787109375,
0.00646209716796875,
0.0038738250732421875,
0.07080078125,
-0.05413818359375,
0.067138671875,
0.04248046875,
-0.049591064453125,
-0.07635498046875,
-0.017120361328125,
-0.007099151611328125,
-0.08203125,
0.04071044921875,
0.01947021484375,
0.01087188720703125,
0.0037441253662109375,
-0.0643310546875,
-0.080078125,
0.09478759765625,
0.028289794921875,
-0.040130615234375,
-0.003902435302734375,
0.006816864013671875,
0.045928955078125,
-0.0286712646484375,
0.03814697265625,
0.0484619140625,
0.03582763671875,
-0.00469207763671875,
-0.0888671875,
0.0256195068359375,
-0.0276947021484375,
0.015106201171875,
-0.00925445556640625,
-0.0814208984375,
0.07855224609375,
-0.037200927734375,
-0.00640106201171875,
0.0294647216796875,
0.051483154296875,
0.037017822265625,
0.012847900390625,
0.0255279541015625,
0.046295166015625,
0.0445556640625,
-0.00476837158203125,
0.0782470703125,
-0.033843994140625,
0.043212890625,
0.04705810546875,
-0.00543212890625,
0.05426025390625,
0.024688720703125,
-0.032867431640625,
0.05413818359375,
0.058074951171875,
-0.0190582275390625,
0.0244598388671875,
0.024169921875,
-0.000006139278411865234,
-0.00875091552734375,
0.00457000732421875,
-0.058685302734375,
0.032073974609375,
0.02740478515625,
-0.02716064453125,
0.0014009475708007812,
-0.0105743408203125,
0.021759033203125,
-0.00901031494140625,
-0.00457763671875,
0.049896240234375,
0.01239776611328125,
-0.046905517578125,
0.08740234375,
0.009124755859375,
0.0687255859375,
-0.03717041015625,
-0.005859375,
-0.0287933349609375,
0.00841522216796875,
-0.0447998046875,
-0.04412841796875,
0.0154266357421875,
0.013763427734375,
-0.00185394287109375,
-0.00952911376953125,
0.035797119140625,
-0.01538848876953125,
-0.037567138671875,
0.02850341796875,
0.0135955810546875,
0.0175628662109375,
0.007740020751953125,
-0.058380126953125,
0.033660888671875,
0.0187835693359375,
-0.03131103515625,
0.0198974609375,
0.0159759521484375,
0.0117034912109375,
0.06439208984375,
0.05474853515625,
-0.0019779205322265625,
0.0074615478515625,
-0.0149383544921875,
0.08599853515625,
-0.05682373046875,
-0.03253173828125,
-0.0640869140625,
0.0445556640625,
0.01153564453125,
-0.035247802734375,
0.048614501953125,
0.032806396484375,
0.0618896484375,
-0.007843017578125,
0.05328369140625,
-0.021728515625,
0.00980377197265625,
-0.041229248046875,
0.05340576171875,
-0.050079345703125,
0.0279541015625,
-0.042144775390625,
-0.07159423828125,
-0.0199432373046875,
0.0677490234375,
-0.01049041748046875,
0.0088348388671875,
0.047454833984375,
0.082763671875,
0.016326904296875,
-0.0021457672119140625,
0.01100921630859375,
0.02178955078125,
0.03363037109375,
0.07073974609375,
0.066650390625,
-0.04766845703125,
0.051177978515625,
-0.04498291015625,
-0.022735595703125,
-0.02349853515625,
-0.07171630859375,
-0.07269287109375,
-0.042327880859375,
-0.032196044921875,
-0.035552978515625,
-0.0155792236328125,
0.0728759765625,
0.053802490234375,
-0.047698974609375,
-0.035552978515625,
-0.00786590576171875,
0.0286865234375,
-0.013916015625,
-0.0179290771484375,
0.02685546875,
-0.0160369873046875,
-0.057373046875,
0.0191497802734375,
0.001399993896484375,
0.007755279541015625,
-0.0177764892578125,
-0.015960693359375,
-0.01152801513671875,
-0.00011587142944335938,
0.034881591796875,
0.02642822265625,
-0.060333251953125,
-0.0188446044921875,
0.00351715087890625,
-0.0161285400390625,
0.01446533203125,
0.03271484375,
-0.050048828125,
0.0021724700927734375,
0.0297088623046875,
0.0266876220703125,
0.029052734375,
-0.01439666748046875,
0.017791748046875,
-0.0291900634765625,
0.02618408203125,
0.00033783912658691406,
0.03631591796875,
0.0124359130859375,
-0.0439453125,
0.054931640625,
0.023406982421875,
-0.052642822265625,
-0.06414794921875,
0.00521087646484375,
-0.0823974609375,
-0.0176239013671875,
0.0966796875,
-0.01305389404296875,
-0.02679443359375,
0.010284423828125,
-0.0316162109375,
0.0168914794921875,
-0.035980224609375,
0.055755615234375,
0.0275726318359375,
-0.00856781005859375,
-0.004306793212890625,
-0.0255889892578125,
0.02459716796875,
0.0269927978515625,
-0.07366943359375,
-0.0091094970703125,
0.02642822265625,
0.032073974609375,
0.01457977294921875,
0.05340576171875,
-0.0081329345703125,
0.0162811279296875,
-0.0013875961303710938,
0.02728271484375,
-0.007030487060546875,
-0.016204833984375,
-0.0267791748046875,
-0.0008406639099121094,
-0.0112457275390625,
-0.004901885986328125
]
] |
FreedomIntelligence/phoenix-inst-chat-7b | 2023-05-31T09:12:13.000Z | [
"transformers",
"pytorch",
"bloom",
"text-generation",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | FreedomIntelligence | null | null | FreedomIntelligence/phoenix-inst-chat-7b | 44 | 7,484 | transformers | 2023-04-11T13:39:18 | ---
license: apache-2.0
---
Please see our [LLMZoo](https://github.com/FreedomIntelligence/LLMZoo) project: https://github.com/FreedomIntelligence/LLMZoo. | 154 | [
[
-0.045623779296875,
-0.00772857666015625,
0.04022216796875,
0.029754638671875,
-0.01439666748046875,
0.006221771240234375,
0.0086669921875,
-0.016510009765625,
0.04248046875,
0.035003662109375,
-0.05975341796875,
-0.057525634765625,
-0.04302978515625,
-0.0181884765625,
-0.0262908935546875,
0.03253173828125,
0.00433349609375,
0.01995849609375,
-0.0233917236328125,
-0.0214080810546875,
0.01204681396484375,
-0.01447296142578125,
-0.03033447265625,
-0.0285491943359375,
0.048095703125,
0.08294677734375,
0.039154052734375,
0.0106964111328125,
0.08514404296875,
0.0133514404296875,
0.002918243408203125,
-0.0238189697265625,
-0.032440185546875,
-0.0036754608154296875,
-0.01116180419921875,
-0.019805908203125,
-0.035858154296875,
0.0249176025390625,
0.0301361083984375,
0.055023193359375,
-0.01055908203125,
0.03790283203125,
-0.01418304443359375,
0.081787109375,
-0.0426025390625,
0.0179290771484375,
-0.0153350830078125,
0.00006836652755737305,
-0.0115509033203125,
-0.009490966796875,
0.003047943115234375,
-0.0469970703125,
-0.0182037353515625,
-0.06610107421875,
0.002197265625,
0.034088134765625,
0.050018310546875,
0.03155517578125,
-0.015899658203125,
-0.0155181884765625,
-0.0263824462890625,
0.0255126953125,
-0.0159454345703125,
0.0159454345703125,
0.0311279296875,
0.059906005859375,
-0.01427459716796875,
-0.05548095703125,
-0.0262908935546875,
-0.03485107421875,
0.0156402587890625,
0.0294647216796875,
-0.006649017333984375,
-0.006855010986328125,
0.00677490234375,
0.026153564453125,
-0.055816650390625,
-0.00887298583984375,
-0.039276123046875,
-0.0206756591796875,
0.03778076171875,
0.0137481689453125,
0.0093841552734375,
0.0151824951171875,
-0.025115966796875,
0.01450347900390625,
-0.04974365234375,
-0.0031452178955078125,
0.02996826171875,
-0.0015058517456054688,
-0.0259857177734375,
0.073974609375,
-0.05230712890625,
0.008392333984375,
-0.00970458984375,
0.02923583984375,
0.036224365234375,
-0.0103302001953125,
-0.0262451171875,
-0.0194244384765625,
0.0251922607421875,
0.054931640625,
0.01474761962890625,
-0.00460052490234375,
-0.0272369384765625,
-0.000014543533325195312,
0.0077056884765625,
-0.0657958984375,
-0.038421630859375,
0.00600433349609375,
-0.052093505859375,
-0.01256561279296875,
0.0134735107421875,
-0.03558349609375,
-0.02362060546875,
-0.01284027099609375,
0.03155517578125,
-0.0186920166015625,
-0.0340576171875,
0.0077362060546875,
-0.0278472900390625,
0.0211944580078125,
0.0185394287109375,
-0.057098388671875,
0.01357269287109375,
0.06585693359375,
0.07952880859375,
0.003185272216796875,
-0.032196044921875,
-0.045867919921875,
-0.007724761962890625,
-0.0033321380615234375,
0.06939697265625,
-0.01351165771484375,
-0.071533203125,
0.000044286251068115234,
0.013427734375,
0.038330078125,
-0.029754638671875,
0.022247314453125,
-0.03729248046875,
-0.012664794921875,
-0.01678466796875,
-0.03497314453125,
-0.0270843505859375,
0.048309326171875,
-0.059326171875,
0.039398193359375,
0.011627197265625,
-0.056243896484375,
0.034271240234375,
-0.091552734375,
0.0026912689208984375,
0.02734375,
-0.021514892578125,
-0.04144287109375,
-0.01328277587890625,
-0.01232147216796875,
0.00978851318359375,
0.0017375946044921875,
0.0092315673828125,
-0.050811767578125,
-0.029022216796875,
0.015045166015625,
0.0023975372314453125,
0.0850830078125,
0.026763916015625,
-0.0007047653198242188,
0.044036865234375,
-0.0516357421875,
-0.01294708251953125,
-0.00661468505859375,
-0.031036376953125,
-0.001708984375,
-0.0161590576171875,
0.0088958740234375,
-0.0291290283203125,
0.04901123046875,
-0.03533935546875,
0.04656982421875,
-0.00949859619140625,
0.020172119140625,
0.0294952392578125,
0.0033168792724609375,
0.0584716796875,
-0.0251617431640625,
0.032440185546875,
-0.0067901611328125,
0.051849365234375,
-0.0015354156494140625,
-0.051483154296875,
-0.06768798828125,
-0.0237579345703125,
0.0075225830078125,
0.050750732421875,
-0.03509521484375,
0.0249481201171875,
-0.0252838134765625,
-0.040924072265625,
0.0101470947265625,
0.0135345458984375,
-0.010284423828125,
0.00914764404296875,
-0.006259918212890625,
-0.048797607421875,
-0.02801513671875,
-0.08734130859375,
0.01021575927734375,
-0.0285797119140625,
-0.049102783203125,
0.01546478271484375,
0.07080078125,
-0.045745849609375,
0.06915283203125,
-0.043182373046875,
0.0016984939575195312,
0.0037326812744140625,
0.0174560546875,
0.0215606689453125,
0.0194091796875,
0.0440673828125,
-0.054840087890625,
-0.019073486328125,
-0.0245513916015625,
-0.01202392578125,
-0.0083465576171875,
0.01480865478515625,
-0.0450439453125,
-0.026275634765625,
0.0016508102416992188,
-0.05908203125,
0.04754638671875,
0.0178985595703125,
-0.05767822265625,
0.0291900634765625,
0.018463134765625,
0.0101776123046875,
-0.0848388671875,
0.00998687744140625,
-0.0167083740234375,
-0.0262908935546875,
-0.014190673828125,
0.0120086669921875,
0.03106689453125,
-0.0119476318359375,
-0.035308837890625,
-0.0018854141235351562,
-0.043365478515625,
-0.038482666015625,
0.0009469985961914062,
0.004413604736328125,
0.001209259033203125,
-0.001953125,
-0.0126495361328125,
0.06951904296875,
0.05908203125,
-0.0124359130859375,
0.06103515625,
0.0153045654296875,
-0.0219573974609375,
0.0099639892578125,
-0.01678466796875,
-0.0017671585083007812,
0.0020961761474609375,
0.00974273681640625,
-0.055816650390625,
-0.0416259765625,
0.040557861328125,
-0.03240966796875,
0.004055023193359375,
-0.013702392578125,
-0.027435302734375,
-0.03924560546875,
-0.07763671875,
0.005157470703125,
0.0321044921875,
-0.053924560546875,
0.006610870361328125,
0.043426513671875,
-0.002124786376953125,
-0.038482666015625,
-0.06982421875,
0.023468017578125,
-0.0221405029296875,
-0.027801513671875,
0.012847900390625,
-0.02484130859375,
-0.04400634765625,
0.0030727386474609375,
0.017578125,
-0.02099609375,
-0.00001329183578491211,
0.063232421875,
0.03399658203125,
-0.0272674560546875,
0.0105133056640625,
-0.038360595703125,
-0.0180206298828125,
0.01207733154296875,
-0.02178955078125,
0.01328277587890625,
-0.053466796875,
-0.03802490234375,
-0.0147705078125,
0.03369140625,
0.046905517578125,
-0.0012159347534179688,
0.033233642578125,
0.032257080078125,
-0.023773193359375,
0.0168304443359375,
-0.038604736328125,
-0.003932952880859375,
-0.0377197265625,
-0.0050811767578125,
-0.0091552734375,
-0.06781005859375,
0.026458740234375,
-0.034027099609375,
0.0008025169372558594,
0.0304412841796875,
0.058807373046875,
-0.0110321044921875,
0.018310546875,
0.08880615234375,
-0.035736083984375,
0.031524658203125,
-0.043426513671875,
0.0113983154296875,
-0.0251922607421875,
-0.018035888671875,
-0.027069091796875,
-0.02532958984375,
-0.0231781005859375,
-0.0330810546875,
-0.0031452178955078125,
0.0189971923828125,
-0.041168212890625,
0.02392578125,
-0.043182373046875,
0.057159423828125,
0.0648193359375,
-0.0294952392578125,
0.0369873046875,
0.00811004638671875,
0.018035888671875,
0.0138397216796875,
-0.01070404052734375,
-0.052337646484375,
0.0985107421875,
0.022796630859375,
0.0625,
0.00711822509765625,
0.076171875,
0.0251617431640625,
0.053802490234375,
-0.08135986328125,
0.047332763671875,
0.0184173583984375,
-0.0316162109375,
-0.0282745361328125,
-0.006099700927734375,
-0.06390380859375,
-0.00783538818359375,
-0.01287841796875,
-0.0160675048828125,
0.01140594482421875,
0.003047943115234375,
-0.02581787109375,
0.020904541015625,
-0.039215087890625,
0.046234130859375,
-0.025115966796875,
0.01432037353515625,
-0.0242919921875,
-0.0256805419921875,
0.03863525390625,
0.0186614990234375,
0.0030193328857421875,
-0.03607177734375,
-0.032012939453125,
0.0308685302734375,
-0.03057861328125,
0.061004638671875,
-0.0081939697265625,
0.0102996826171875,
0.0313720703125,
-0.00820159912109375,
-0.00759124755859375,
0.056365966796875,
-0.0012950897216796875,
0.0218353271484375,
-0.0297393798828125,
-0.02020263671875,
0.00024819374084472656,
0.0655517578125,
-0.04400634765625,
-0.0288543701171875,
-0.02362060546875,
-0.06011962890625,
0.0180206298828125,
0.03363037109375,
0.00890350341796875,
0.00513458251953125,
-0.0262908935546875,
0.01482391357421875,
0.026947021484375,
0.0192413330078125,
0.0408935546875,
0.049835205078125,
-0.049102783203125,
-0.024871826171875,
0.04248046875,
-0.00226593017578125,
-0.0177459716796875,
0.0276641845703125,
0.003688812255859375,
-0.0157928466796875,
-0.07098388671875,
-0.035614013671875,
0.0102691650390625,
-0.04351806640625,
-0.032012939453125,
-0.00006395578384399414,
-0.0098114013671875,
-0.048614501953125,
-0.025115966796875,
-0.00856781005859375,
-0.072998046875,
-0.0377197265625,
-0.006916046142578125,
0.062744140625,
0.04669189453125,
-0.019378662109375,
0.0244598388671875,
-0.048675537109375,
0.0404052734375,
0.020599365234375,
0.044281005859375,
-0.0718994140625,
-0.0023250579833984375,
-0.0239410400390625,
-0.021087646484375,
-0.056884765625,
-0.03131103515625,
0.03594970703125,
0.0186920166015625,
0.04010009765625,
0.046112060546875,
-0.005519866943359375,
0.024017333984375,
-0.041778564453125,
0.07562255859375,
0.038330078125,
-0.043914794921875,
0.058441162109375,
-0.0135498046875,
0.0328369140625,
0.06939697265625,
0.0426025390625,
-0.0155029296875,
-0.022247314453125,
-0.0293731689453125,
-0.041290283203125,
-0.00988006591796875,
0.0125579833984375,
-0.007091522216796875,
0.00934600830078125,
0.02886962890625,
0.01049041748046875,
0.01171112060546875,
-0.056854248046875,
-0.01058197021484375,
-0.0254974365234375,
-0.011260986328125,
0.004207611083984375,
-0.018157958984375,
-0.03802490234375,
-0.0011014938354492188,
0.0313720703125,
-0.0011816024780273438,
0.005420684814453125,
0.015716552734375,
-0.00609588623046875,
-0.0157012939453125,
0.01068115234375,
0.036376953125,
0.056884765625,
-0.0254974365234375,
0.00923919677734375,
0.00913238525390625,
-0.06573486328125,
-0.00695037841796875,
0.003875732421875,
-0.025115966796875,
0.0078277587890625,
0.03564453125,
0.0136260986328125,
-0.0162811279296875,
-0.043670654296875,
0.0489501953125,
-0.0083160400390625,
-0.045257568359375,
-0.055816650390625,
0.004978179931640625,
0.056060791015625,
0.05645751953125,
0.0156402587890625,
0.017333984375,
0.05413818359375,
-0.0186614990234375,
0.01763916015625,
0.03985595703125,
0.002590179443359375,
0.004436492919921875,
0.039398193359375,
0.035003662109375,
-0.058807373046875,
0.0361328125,
-0.026611328125,
-0.041961669921875,
0.06451416015625,
0.01215362548828125,
0.06646728515625,
-0.01065826416015625,
0.0011157989501953125,
0.01274871826171875,
0.056365966796875,
0.0043182373046875,
0.042724609375,
-0.030426025390625,
-0.01131439208984375,
0.00339508056640625,
-0.0176544189453125,
-0.066650390625,
-0.013885498046875,
-0.0301055908203125,
0.029937744140625,
-0.041351318359375,
-0.0175628662109375,
-0.0237579345703125,
0.0204315185546875,
-0.033416748046875,
-0.0007205009460449219,
0.03179931640625,
0.11505126953125,
-0.0223541259765625,
0.08544921875,
0.058624267578125,
-0.0059051513671875,
-0.012664794921875,
-0.01227569580078125,
0.0089263916015625,
-0.049835205078125,
0.06036376953125,
0.06024169921875,
-0.0548095703125,
-0.0293731689453125,
-0.0426025390625,
-0.0648193359375,
0.123779296875,
0.00627899169921875,
-0.0750732421875,
0.0196075439453125,
0.0098419189453125,
0.01000213623046875,
-0.03277587890625,
-0.006916046142578125,
0.0179595947265625,
0.037750244140625,
-0.01557159423828125,
-0.0465087890625,
-0.0254364013671875,
-0.03570556640625,
0.0018968582153320312,
0.01427459716796875,
-0.04656982421875,
0.032501220703125,
-0.0223236083984375,
-0.01580810546875,
0.003631591796875,
0.05560302734375,
-0.01264190673828125,
0.054840087890625,
0.048675537109375,
0.062286376953125,
0.031951904296875,
0.006015777587890625,
0.03436279296875,
-0.0186767578125,
0.048095703125,
0.07562255859375,
-0.020538330078125,
0.065185546875,
0.06060791015625,
-0.015838623046875,
0.0013875961303710938,
0.041351318359375,
-0.0249481201171875,
0.053436279296875,
0.025238037109375,
-0.0390625,
0.0086822509765625,
0.0286712646484375,
-0.0291290283203125,
-0.0036754608154296875,
0.03839111328125,
-0.004810333251953125,
-0.00382232666015625,
0.0088043212890625,
-0.032623291015625,
-0.0264739990234375,
-0.029754638671875,
0.0197906494140625,
0.005626678466796875,
-0.0092620849609375,
0.0146484375,
-0.0050201416015625,
0.06500244140625,
-0.03582763671875,
-0.0011539459228515625,
-0.023773193359375,
-0.005229949951171875,
-0.0244293212890625,
-0.09173583984375,
0.056884765625,
0.00409698486328125,
0.004680633544921875,
-0.0017719268798828125,
0.047393798828125,
-0.00952911376953125,
-0.054534912109375,
0.03668212890625,
0.003215789794921875,
0.017486572265625,
0.005687713623046875,
-0.0426025390625,
-0.0030193328857421875,
-0.0010251998901367188,
-0.051513671875,
-0.0009570121765136719,
0.0008912086486816406,
0.0193023681640625,
0.0670166015625,
0.045135498046875,
0.007099151611328125,
-0.0081024169921875,
0.0281982421875,
0.072265625,
-0.050537109375,
-0.0172271728515625,
-0.051483154296875,
0.0245513916015625,
-0.033721923828125,
-0.01125335693359375,
0.05145263671875,
0.0848388671875,
0.060699462890625,
-0.0404052734375,
0.03204345703125,
0.020751953125,
0.029632568359375,
-0.004344940185546875,
0.07293701171875,
-0.03564453125,
0.0035190582275390625,
0.04449462890625,
-0.06451416015625,
-0.075927734375,
0.041748046875,
-0.0090179443359375,
-0.0513916015625,
0.029327392578125,
0.08172607421875,
-0.01611328125,
0.007328033447265625,
0.0362548828125,
-0.01172637939453125,
0.00799560546875,
0.039093017578125,
0.07989501953125,
0.01373291015625,
0.061553955078125,
-0.04107666015625,
-0.050201416015625,
-0.038543701171875,
-0.07537841796875,
-0.061004638671875,
-0.0223541259765625,
-0.0169525146484375,
0.01256561279296875,
-0.008209228515625,
0.045928955078125,
0.053253173828125,
-0.04559326171875,
-0.026214599609375,
-0.01061248779296875,
-0.004398345947265625,
0.0182647705078125,
-0.0122528076171875,
-0.00719451904296875,
-0.002147674560546875,
-0.0509033203125,
0.019744873046875,
0.056396484375,
0.0049285888671875,
-0.0207366943359375,
-0.00742340087890625,
0.00521087646484375,
-0.001270294189453125,
0.038482666015625,
0.0157623291015625,
-0.0457763671875,
-0.017181396484375,
-0.01013946533203125,
-0.01447296142578125,
0.0246124267578125,
0.072509765625,
-0.0254364013671875,
-0.0011234283447265625,
0.038330078125,
0.04388427734375,
0.03466796875,
-0.017974853515625,
0.0014257431030273438,
0.0045166015625,
0.02325439453125,
-0.002101898193359375,
0.053924560546875,
0.0231475830078125,
-0.04351806640625,
0.068603515625,
0.0053253173828125,
-0.07183837890625,
-0.0419921875,
0.032562255859375,
-0.0980224609375,
0.01009368896484375,
0.025604248046875,
-0.0196380615234375,
-0.0202789306640625,
0.00673675537109375,
-0.043914794921875,
0.0205230712890625,
-0.0289306640625,
0.060516357421875,
0.047821044921875,
-0.0192108154296875,
0.0095672607421875,
-0.049407958984375,
-0.0011587142944335938,
0.000016748905181884766,
-0.0782470703125,
-0.02288818359375,
0.043914794921875,
0.0245513916015625,
0.0268096923828125,
0.036529541015625,
-0.01438140869140625,
0.029632568359375,
0.018798828125,
0.0325927734375,
-0.01543426513671875,
-0.0269012451171875,
-0.0294647216796875,
-0.00403594970703125,
0.0158538818359375,
-0.06280517578125
]
] |
flaubert/flaubert_small_cased | 2021-05-19T16:56:07.000Z | [
"transformers",
"pytorch",
"flaubert",
"fill-mask",
"bert",
"language-model",
"flue",
"french",
"flaubert-small",
"cased",
"fr",
"dataset:flaubert",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | fill-mask | flaubert | null | null | flaubert/flaubert_small_cased | 1 | 7,474 | transformers | 2022-03-02T23:29:05 | ---
language: fr
license: mit
datasets:
- flaubert
metrics:
- flue
tags:
- bert
- language-model
- flaubert
- flue
- french
- flaubert-small
- cased
---
# FlauBERT: Unsupervised Language Model Pre-training for French
**FlauBERT** is a French BERT trained on a very large and heterogeneous French corpus. Models of different sizes are trained using the new CNRS (French National Centre for Scientific Research) [Jean Zay](http://www.idris.fr/eng/jean-zay/ ) supercomputer.
Along with FlauBERT comes [**FLUE**](https://github.com/getalp/Flaubert/tree/master/flue): an evaluation setup for French NLP systems similar to the popular GLUE benchmark. The goal is to enable further reproducible experiments in the future and to share models and progress on the French language.For more details please refer to the [official website](https://github.com/getalp/Flaubert).
## FlauBERT models
| Model name | Number of layers | Attention Heads | Embedding Dimension | Total Parameters |
| :------: | :---: | :---: | :---: | :---: |
| `flaubert-small-cased` | 6 | 8 | 512 | 54 M |
| `flaubert-base-uncased` | 12 | 12 | 768 | 137 M |
| `flaubert-base-cased` | 12 | 12 | 768 | 138 M |
| `flaubert-large-cased` | 24 | 16 | 1024 | 373 M |
**Note:** `flaubert-small-cased` is partially trained so performance is not guaranteed. Consider using it for debugging purpose only.
## Using FlauBERT with Hugging Face's Transformers
```python
import torch
from transformers import FlaubertModel, FlaubertTokenizer
# Choose among ['flaubert/flaubert_small_cased', 'flaubert/flaubert_base_uncased',
# 'flaubert/flaubert_base_cased', 'flaubert/flaubert_large_cased']
modelname = 'flaubert/flaubert_base_cased'
# Load pretrained model and tokenizer
flaubert, log = FlaubertModel.from_pretrained(modelname, output_loading_info=True)
flaubert_tokenizer = FlaubertTokenizer.from_pretrained(modelname, do_lowercase=False)
# do_lowercase=False if using cased models, True if using uncased ones
sentence = "Le chat mange une pomme."
token_ids = torch.tensor([flaubert_tokenizer.encode(sentence)])
last_layer = flaubert(token_ids)[0]
print(last_layer.shape)
# torch.Size([1, 8, 768]) -> (batch size x number of tokens x embedding dimension)
# The BERT [CLS] token correspond to the first hidden state of the last layer
cls_embedding = last_layer[:, 0, :]
```
**Notes:** if your `transformers` version is <=2.10.0, `modelname` should take one
of the following values:
```
['flaubert-small-cased', 'flaubert-base-uncased', 'flaubert-base-cased', 'flaubert-large-cased']
```
## References
If you use FlauBERT or the FLUE Benchmark for your scientific publication, or if you find the resources in this repository useful, please cite one of the following papers:
[LREC paper](http://www.lrec-conf.org/proceedings/lrec2020/pdf/2020.lrec-1.302.pdf)
```
@InProceedings{le2020flaubert,
author = {Le, Hang and Vial, Lo\"{i}c and Frej, Jibril and Segonne, Vincent and Coavoux, Maximin and Lecouteux, Benjamin and Allauzen, Alexandre and Crabb\'{e}, Beno\^{i}t and Besacier, Laurent and Schwab, Didier},
title = {FlauBERT: Unsupervised Language Model Pre-training for French},
booktitle = {Proceedings of The 12th Language Resources and Evaluation Conference},
month = {May},
year = {2020},
address = {Marseille, France},
publisher = {European Language Resources Association},
pages = {2479--2490},
url = {https://www.aclweb.org/anthology/2020.lrec-1.302}
}
```
[TALN paper](https://hal.archives-ouvertes.fr/hal-02784776/)
```
@inproceedings{le2020flaubert,
title = {FlauBERT: des mod{\`e}les de langue contextualis{\'e}s pr{\'e}-entra{\^\i}n{\'e}s pour le fran{\c{c}}ais},
author = {Le, Hang and Vial, Lo{\"\i}c and Frej, Jibril and Segonne, Vincent and Coavoux, Maximin and Lecouteux, Benjamin and Allauzen, Alexandre and Crabb{\'e}, Beno{\^\i}t and Besacier, Laurent and Schwab, Didier},
booktitle = {Actes de la 6e conf{\'e}rence conjointe Journ{\'e}es d'{\'E}tudes sur la Parole (JEP, 31e {\'e}dition), Traitement Automatique des Langues Naturelles (TALN, 27e {\'e}dition), Rencontre des {\'E}tudiants Chercheurs en Informatique pour le Traitement Automatique des Langues (R{\'E}CITAL, 22e {\'e}dition). Volume 2: Traitement Automatique des Langues Naturelles},
pages = {268--278},
year = {2020},
organization = {ATALA}
}
``` | 4,472 | [
[
-0.0251922607421875,
-0.055450439453125,
0.0262451171875,
0.01352691650390625,
-0.00099945068359375,
0.00377655029296875,
-0.0203399658203125,
-0.007904052734375,
0.0252227783203125,
0.03680419921875,
-0.030914306640625,
-0.035552978515625,
-0.04705810546875,
-0.0139312744140625,
-0.044403076171875,
0.061492919921875,
-0.01396942138671875,
-0.002414703369140625,
-0.0051422119140625,
-0.0025043487548828125,
0.0103912353515625,
-0.05877685546875,
-0.037109375,
-0.017791748046875,
0.036468505859375,
0.005107879638671875,
0.040435791015625,
0.022125244140625,
0.0115203857421875,
0.031280517578125,
-0.0103607177734375,
-0.005161285400390625,
-0.041900634765625,
-0.0167388916015625,
0.006732940673828125,
-0.0253143310546875,
-0.0255126953125,
-0.00397491455078125,
0.05230712890625,
0.042388916015625,
0.009429931640625,
0.0010213851928710938,
0.00007510185241699219,
0.051422119140625,
-0.034912109375,
0.022674560546875,
-0.026824951171875,
0.01299285888671875,
-0.00969696044921875,
-0.002872467041015625,
-0.04168701171875,
-0.01006317138671875,
0.0226593017578125,
-0.0301361083984375,
0.0258331298828125,
-0.008697509765625,
0.08551025390625,
0.005481719970703125,
-0.032684326171875,
0.0031642913818359375,
-0.0511474609375,
0.06549072265625,
-0.050628662109375,
0.046295166015625,
0.01258087158203125,
0.0003437995910644531,
-0.02001953125,
-0.0833740234375,
-0.04949951171875,
-0.0128021240234375,
-0.01503753662109375,
0.004039764404296875,
-0.01824951171875,
-0.005168914794921875,
0.019927978515625,
0.027496337890625,
-0.037017822265625,
-0.0139007568359375,
-0.0423583984375,
-0.037109375,
0.050811767578125,
-0.0168609619140625,
0.017578125,
-0.0006227493286132812,
-0.03643798828125,
-0.038909912109375,
-0.0185089111328125,
0.018157958984375,
0.0255279541015625,
0.015472412109375,
-0.02532958984375,
0.033416748046875,
-0.000038623809814453125,
0.035858154296875,
-0.0031757354736328125,
0.005496978759765625,
0.060211181640625,
-0.011016845703125,
-0.0223846435546875,
-0.0035305023193359375,
0.084228515625,
0.005901336669921875,
0.03460693359375,
-0.0092926025390625,
-0.02618408203125,
-0.0202789306640625,
0.01084136962890625,
-0.05810546875,
-0.03009033203125,
0.0239105224609375,
-0.03564453125,
-0.0231170654296875,
0.02294921875,
-0.042236328125,
-0.0030364990234375,
-0.00423431396484375,
0.05902099609375,
-0.041534423828125,
-0.034271240234375,
0.0103302001953125,
0.00276947021484375,
0.031982421875,
0.004596710205078125,
-0.06732177734375,
0.0132904052734375,
0.03216552734375,
0.0618896484375,
0.0020389556884765625,
-0.0310821533203125,
-0.032012939453125,
-0.012298583984375,
-0.0216217041015625,
0.033935546875,
-0.031707763671875,
-0.0149078369140625,
0.0214996337890625,
0.0234375,
-0.0290069580078125,
-0.015106201171875,
0.0552978515625,
-0.0286102294921875,
0.0233154296875,
-0.01309967041015625,
-0.046051025390625,
-0.0302886962890625,
-0.01361846923828125,
-0.049346923828125,
0.0849609375,
0.0341796875,
-0.06280517578125,
0.01111602783203125,
-0.03887939453125,
-0.0305938720703125,
-0.0094451904296875,
-0.0165557861328125,
-0.031585693359375,
0.01678466796875,
0.0301971435546875,
0.0474853515625,
0.0013818740844726562,
-0.004596710205078125,
-0.00786590576171875,
-0.022125244140625,
0.0211029052734375,
-0.018798828125,
0.0780029296875,
0.0127105712890625,
-0.02734375,
0.0187225341796875,
-0.0523681640625,
0.00997161865234375,
0.018798828125,
-0.0158538818359375,
0.0004439353942871094,
-0.0137176513671875,
0.022247314453125,
0.010894775390625,
0.0384521484375,
-0.043426513671875,
0.01024627685546875,
-0.03680419921875,
0.04779052734375,
0.05548095703125,
0.0064849853515625,
0.026275634765625,
-0.0225067138671875,
0.021484375,
0.0225982666015625,
0.021209716796875,
-0.0024547576904296875,
-0.0281982421875,
-0.07977294921875,
-0.027801513671875,
0.042022705078125,
0.044464111328125,
-0.04150390625,
0.05694580078125,
-0.01522064208984375,
-0.03839111328125,
-0.02587890625,
-0.004093170166015625,
0.013824462890625,
0.0174713134765625,
0.03594970703125,
-0.01226806640625,
-0.026336669921875,
-0.08349609375,
-0.00997161865234375,
-0.00039386749267578125,
0.004150390625,
-0.00359344482421875,
0.052886962890625,
-0.03021240234375,
0.047393798828125,
-0.0287322998046875,
-0.02960205078125,
-0.0184783935546875,
0.01213836669921875,
0.032562255859375,
0.0491943359375,
0.06414794921875,
-0.037322998046875,
-0.037445068359375,
-0.0143890380859375,
-0.045684814453125,
0.021820068359375,
-0.005138397216796875,
-0.026031494140625,
0.028167724609375,
0.031494140625,
-0.0361328125,
0.038055419921875,
0.0269622802734375,
-0.02447509765625,
0.025299072265625,
-0.0207672119140625,
0.004169464111328125,
-0.0703125,
-0.007678985595703125,
-0.0008559226989746094,
-0.0186614990234375,
-0.057098388671875,
-0.005924224853515625,
0.005954742431640625,
0.00658416748046875,
-0.045013427734375,
0.03912353515625,
-0.03570556640625,
0.009796142578125,
0.01470184326171875,
0.0139007568359375,
-0.007289886474609375,
0.0628662109375,
0.0062103271484375,
0.031890869140625,
0.07159423828125,
-0.032562255859375,
0.0195159912109375,
0.027130126953125,
-0.033538818359375,
0.009765625,
-0.05126953125,
0.01450347900390625,
-0.001850128173828125,
0.0167236328125,
-0.05841064453125,
-0.0020809173583984375,
0.01556396484375,
-0.03436279296875,
0.0283355712890625,
-0.0005745887756347656,
-0.059661865234375,
-0.033447265625,
-0.026336669921875,
0.0221099853515625,
0.047943115234375,
-0.033233642578125,
0.05047607421875,
0.015869140625,
0.008697509765625,
-0.054107666015625,
-0.0718994140625,
-0.026336669921875,
-0.005260467529296875,
-0.0606689453125,
0.02606201171875,
-0.01172637939453125,
0.01126861572265625,
-0.003814697265625,
-0.007396697998046875,
-0.01395416259765625,
-0.0140533447265625,
0.00803375244140625,
-0.000039577484130859375,
-0.021240234375,
-0.0031528472900390625,
0.004215240478515625,
-0.003337860107421875,
0.007007598876953125,
-0.02484130859375,
0.055999755859375,
-0.040924072265625,
-0.022674560546875,
-0.037353515625,
0.01727294921875,
0.047637939453125,
-0.028778076171875,
0.07489013671875,
0.08197021484375,
-0.043731689453125,
-0.01226806640625,
-0.0364990234375,
-0.0248260498046875,
-0.03912353515625,
0.0265655517578125,
-0.0299072265625,
-0.058013916015625,
0.05352783203125,
0.0186767578125,
0.0121002197265625,
0.051116943359375,
0.041290283203125,
0.0016145706176757812,
0.073486328125,
0.0438232421875,
-0.006519317626953125,
0.0439453125,
-0.06414794921875,
0.0272979736328125,
-0.046295166015625,
-0.02593994140625,
-0.0207366943359375,
-0.03204345703125,
-0.032562255859375,
-0.032867431640625,
0.018310546875,
0.02972412109375,
-0.0249481201171875,
0.03253173828125,
-0.042205810546875,
0.0197601318359375,
0.051422119140625,
0.0217132568359375,
-0.01065826416015625,
0.01959228515625,
-0.045501708984375,
-0.00567626953125,
-0.05535888671875,
-0.036376953125,
0.07806396484375,
0.044158935546875,
0.03436279296875,
0.0159912109375,
0.07177734375,
0.0133514404296875,
0.0011892318725585938,
-0.050628662109375,
0.041748046875,
-0.01451873779296875,
-0.04510498046875,
-0.017974853515625,
-0.02069091796875,
-0.071044921875,
0.0203704833984375,
-0.0144500732421875,
-0.08258056640625,
0.0255584716796875,
0.0021877288818359375,
-0.0258331298828125,
0.0296630859375,
-0.0521240234375,
0.0777587890625,
-0.0289306640625,
-0.023773193359375,
-0.0159759521484375,
-0.039794921875,
0.0197601318359375,
-0.00679779052734375,
0.011322021484375,
0.00858306884765625,
0.01204681396484375,
0.07562255859375,
-0.0450439453125,
0.056488037109375,
-0.01526641845703125,
-0.0035991668701171875,
0.031982421875,
0.006122589111328125,
0.03778076171875,
0.01171112060546875,
-0.0054473876953125,
0.02001953125,
0.0291595458984375,
-0.04052734375,
-0.03558349609375,
0.059783935546875,
-0.075927734375,
-0.028839111328125,
-0.05364990234375,
-0.0252532958984375,
-0.004638671875,
0.0250091552734375,
0.039703369140625,
0.060211181640625,
-0.004047393798828125,
0.0343017578125,
0.049774169921875,
-0.0360107421875,
0.041290283203125,
0.0272369384765625,
-0.0252227783203125,
-0.0199737548828125,
0.06890869140625,
0.01172637939453125,
0.002307891845703125,
0.045135498046875,
0.01396942138671875,
-0.03350830078125,
-0.02593994140625,
-0.005855560302734375,
0.03167724609375,
-0.06536865234375,
0.002902984619140625,
-0.06304931640625,
-0.050933837890625,
-0.029052734375,
-0.0144805908203125,
-0.037506103515625,
-0.040496826171875,
-0.040496826171875,
-0.004169464111328125,
0.03424072265625,
0.0411376953125,
-0.01971435546875,
0.0244598388671875,
-0.05316162109375,
0.002216339111328125,
0.006824493408203125,
0.0200042724609375,
0.005481719970703125,
-0.05767822265625,
-0.0264434814453125,
0.00787353515625,
-0.01358795166015625,
-0.05828857421875,
0.0247955322265625,
0.01450347900390625,
0.067626953125,
0.038055419921875,
0.01861572265625,
0.0196533203125,
-0.030914306640625,
0.06585693359375,
0.01322174072265625,
-0.07470703125,
0.03839111328125,
-0.012115478515625,
0.01166534423828125,
0.034759521484375,
0.033477783203125,
-0.018798828125,
-0.0251922607421875,
-0.07666015625,
-0.0784912109375,
0.056549072265625,
0.0300750732421875,
0.01922607421875,
-0.0100555419921875,
0.01239013671875,
-0.006256103515625,
0.01096343994140625,
-0.0684814453125,
-0.04119873046875,
-0.02960205078125,
-0.0200042724609375,
-0.00586700439453125,
-0.021575927734375,
-0.017822265625,
-0.0426025390625,
0.0625,
0.010101318359375,
0.060882568359375,
0.020477294921875,
-0.02874755859375,
0.004913330078125,
0.0013294219970703125,
0.06982421875,
0.044921875,
-0.0291595458984375,
0.007114410400390625,
0.0169677734375,
-0.0282440185546875,
-0.005657196044921875,
0.0109710693359375,
0.0019855499267578125,
0.00702667236328125,
0.053985595703125,
0.07861328125,
0.01169586181640625,
-0.029815673828125,
0.061737060546875,
-0.004241943359375,
-0.03680419921875,
-0.053985595703125,
0.013824462890625,
-0.01007080078125,
0.035797119140625,
0.033843994140625,
0.006343841552734375,
-0.02447509765625,
-0.0207672119140625,
0.034393310546875,
0.026885986328125,
-0.032440185546875,
-0.0291748046875,
0.057159423828125,
0.006931304931640625,
-0.0230255126953125,
0.039276123046875,
-0.0135498046875,
-0.05499267578125,
0.032379150390625,
0.0227508544921875,
0.075927734375,
-0.005161285400390625,
0.022857666015625,
0.046966552734375,
0.03619384765625,
-0.0019140243530273438,
0.0302734375,
0.00812530517578125,
-0.05633544921875,
-0.0113983154296875,
-0.055694580078125,
0.00946044921875,
0.0411376953125,
-0.04351806640625,
0.0193328857421875,
-0.043731689453125,
-0.0059356689453125,
-0.0087890625,
-0.002353668212890625,
-0.0689697265625,
-0.0009570121765136719,
0.005344390869140625,
0.0738525390625,
-0.06182861328125,
0.07110595703125,
0.04791259765625,
-0.053497314453125,
-0.0655517578125,
0.00762939453125,
-0.0143280029296875,
-0.057708740234375,
0.056915283203125,
0.01125335693359375,
-0.0042724609375,
0.018341064453125,
-0.034820556640625,
-0.0692138671875,
0.0767822265625,
0.0233917236328125,
-0.044769287109375,
0.0142364501953125,
0.0009317398071289062,
0.045135498046875,
-0.0293426513671875,
0.03558349609375,
0.03948974609375,
0.035430908203125,
-0.003353118896484375,
-0.05609130859375,
-0.00241851806640625,
-0.017303466796875,
-0.005275726318359375,
0.0007205009460449219,
-0.06103515625,
0.07806396484375,
-0.01006317138671875,
-0.01519012451171875,
-0.0022125244140625,
0.07159423828125,
-0.0029201507568359375,
-0.0206298828125,
0.040618896484375,
0.03662109375,
0.05010986328125,
-0.0230560302734375,
0.06756591796875,
-0.045623779296875,
0.053741455078125,
0.04986572265625,
0.00807952880859375,
0.0706787109375,
0.0255584716796875,
-0.027496337890625,
0.05621337890625,
0.049163818359375,
-0.00592041015625,
0.050872802734375,
0.0160064697265625,
-0.01207733154296875,
-0.010589599609375,
0.02972412109375,
-0.043609619140625,
0.024749755859375,
0.0295867919921875,
-0.043365478515625,
0.0008783340454101562,
0.0052947998046875,
0.007526397705078125,
-0.00843048095703125,
-0.002323150634765625,
0.0201416015625,
0.00981903076171875,
-0.02587890625,
0.085205078125,
-0.0028896331787109375,
0.039031982421875,
-0.04364013671875,
0.021484375,
-0.0190277099609375,
0.027801513671875,
-0.0207366943359375,
-0.05316162109375,
-0.005298614501953125,
-0.0200653076171875,
-0.0168609619140625,
-0.0092926025390625,
0.0283050537109375,
-0.043426513671875,
-0.054534912109375,
0.035247802734375,
0.0253448486328125,
0.02490234375,
0.01265716552734375,
-0.07073974609375,
0.0131378173828125,
0.01300811767578125,
-0.04583740234375,
0.007076263427734375,
0.0224151611328125,
0.01085662841796875,
0.039276123046875,
0.0274200439453125,
-0.007625579833984375,
0.023895263671875,
0.023956298828125,
0.059112548828125,
-0.031890869140625,
-0.0364990234375,
-0.052154541015625,
0.054595947265625,
0.0011663436889648438,
-0.0194549560546875,
0.03753662109375,
0.050201416015625,
0.0635986328125,
-0.028045654296875,
0.058929443359375,
-0.01255035400390625,
0.027801513671875,
-0.0367431640625,
0.058807373046875,
-0.05731201171875,
0.00960540771484375,
-0.026885986328125,
-0.08184814453125,
-0.0106353759765625,
0.08056640625,
-0.012969970703125,
0.0167999267578125,
0.072265625,
0.0689697265625,
-0.0190887451171875,
-0.015228271484375,
0.0199432373046875,
0.03662109375,
0.0252532958984375,
0.04754638671875,
0.045318603515625,
-0.05487060546875,
0.0347900390625,
-0.04278564453125,
-0.020355224609375,
-0.0283660888671875,
-0.061737060546875,
-0.09423828125,
-0.0772705078125,
-0.041778564453125,
-0.036224365234375,
-0.0196990966796875,
0.068603515625,
0.052886962890625,
-0.0794677734375,
-0.007526397705078125,
-0.0104522705078125,
-0.0099334716796875,
-0.022857666015625,
-0.0211944580078125,
0.050018310546875,
-0.0128173828125,
-0.055999755859375,
0.0279083251953125,
0.0016164779663085938,
0.01532745361328125,
-0.0310821533203125,
-0.0118865966796875,
-0.0347900390625,
0.008544921875,
0.036285400390625,
0.0195159912109375,
-0.062347412109375,
-0.044097900390625,
-0.0140228271484375,
-0.010040283203125,
0.0006041526794433594,
0.046630859375,
-0.044921875,
0.0234375,
0.046051025390625,
0.02679443359375,
0.06597900390625,
-0.0260772705078125,
0.037017822265625,
-0.0762939453125,
0.0399169921875,
0.0097198486328125,
0.044921875,
0.0122528076171875,
-0.0148468017578125,
0.034149169921875,
0.0143280029296875,
-0.04290771484375,
-0.071044921875,
0.01018524169921875,
-0.0772705078125,
-0.0136871337890625,
0.0738525390625,
-0.00801849365234375,
-0.0161590576171875,
0.01505279541015625,
-0.0064849853515625,
0.03564453125,
-0.031829833984375,
0.02471923828125,
0.054107666015625,
-0.01384735107421875,
-0.033660888671875,
-0.057952880859375,
0.0400390625,
0.028656005859375,
-0.03173828125,
-0.019134521484375,
0.0019893646240234375,
0.0186614990234375,
0.02484130859375,
0.0350341796875,
-0.001140594482421875,
-0.0084228515625,
-0.0014858245849609375,
0.008575439453125,
-0.0018949508666992188,
-0.0223236083984375,
-0.005596160888671875,
-0.005947113037109375,
-0.012054443359375,
-0.0187225341796875
]
] |
uukuguy/speechless-codellama-dolphin-orca-platypus-34b | 2023-10-08T08:32:33.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-2",
"code",
"en",
"dataset:ehartford/dolphin",
"dataset:Open-Orca/OpenOrca",
"dataset:garage-bAInd/Open-Platypus",
"arxiv:2308.12950",
"license:llama2",
"model-index",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | uukuguy | null | null | uukuguy/speechless-codellama-dolphin-orca-platypus-34b | 6 | 7,467 | transformers | 2023-09-07T12:07:35 | ---
language:
- en
library_name: transformers
pipeline_tag: text-generation
datasets:
- ehartford/dolphin
- Open-Orca/OpenOrca
- garage-bAInd/Open-Platypus
tags:
- llama-2
- code
license: llama2
model-index:
- name: SpeechlessCoder
results:
- task:
type: text-generation
dataset:
type: openai_humaneval
name: HumanEval
metrics:
- name: pass@1
type: pass@1
value: 0.7012
verified: false
---
<p><h1> speechless-codellama-dolphin-orca-platypus-34b </h1></p>
> 2023.10.06 [uukuguy/speechless-codellama-34b-v2.0](https://huggingface.co/uukuguy/speechless-codellama-34b-v2.0) release. humaneval-python pass@1: 75.61
Fine-tune the Phind/Phind-CodeLlama-34B with Dolphin (1% GPT4), Orca (1% GPT4) and Platypus (100%) datasets.
| Metric | Value |
| --- | --- |
| humaneval-python | 70.12 |
[Big Code Models Leaderboard](https://huggingface.co/spaces/bigcode/bigcode-models-leaderboard)
Phind-CodeLlama-34B-v2: 71.95
WizardCoder-Python-34B-V1.0: 70.73
Phind-CodeLlama-34B-v1: 65.85
WizardCoder-Python-13B-V1.0: 62.19
CodeLlama-34B-Python: 53.29
CodeLlama-34B-Instruct: 50.79
CodeLlama-13B-Instruct: 50.6
CodeLlama-34B: 45.11
CodeLlama-13B-Python: 42.89
CodeLlama-13B: 35.07
[Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
| Metric | Value |
| --- | --- |
| ARC | 52.47 |
| HellaSwag | 74.13 |
| MMLU | 53.47 |
| TruthfulQA | 47.14 |
| Average | 56.80 |
# **Code Llama**
Code Llama is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 34 billion parameters. This is the repository for the base 13B version in the Hugging Face Transformers format. This model is designed for general code synthesis and understanding. Links to other models can be found in the index at the bottom.
| | Base Model | Python | Instruct |
| --- | ----------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- |
| 7B | [codellama/CodeLlama-7b-hf](https://huggingface.co/codellama/CodeLlama-7b-hf) | [codellama/CodeLlama-7b-Python-hf](https://huggingface.co/codellama/CodeLlama-7b-Python-hf) | [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) |
| 13B | [codellama/CodeLlama-13b-hf](https://huggingface.co/codellama/CodeLlama-13b-hf) | [codellama/CodeLlama-13b-Python-hf](https://huggingface.co/codellama/CodeLlama-13b-Python-hf) | [codellama/CodeLlama-13b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-13b-Instruct-hf) |
| 34B | [codellama/CodeLlama-34b-hf](https://huggingface.co/codellama/CodeLlama-34b-hf) | [codellama/CodeLlama-34b-Python-hf](https://huggingface.co/codellama/CodeLlama-34b-Python-hf) | [codellama/CodeLlama-34b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-34b-Instruct-hf) |
## Model Use
To use this model, please make sure to install transformers from `main` until the next version is released:
```bash
pip install git+https://github.com/huggingface/transformers.git@main accelerate
```
Model capabilities:
- [x] Code completion.
- [x] Infilling.
- [ ] Instructions / chat.
- [ ] Python specialist.
```python
from transformers import AutoTokenizer
import transformers
import torch
model = "codellama/CodeLlama-13b-hf"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
sequences = pipeline(
'import socket\n\ndef ping_exponential_backoff(host: str):',
do_sample=True,
top_k=10,
temperature=0.1,
top_p=0.95,
num_return_sequences=1,
eos_token_id=tokenizer.eos_token_id,
max_length=200,
)
for seq in sequences:
print(f"Result: {seq['generated_text']}")
```
## Model Details
*Note: Use of this model is governed by the Meta license. Meta developed and publicly released the Code Llama family of large language models (LLMs).
**Model Developers** Meta
**Variations** Code Llama comes in three model sizes, and three variants:
* Code Llama: base models designed for general code synthesis and understanding
* Code Llama - Python: designed specifically for Python
* Code Llama - Instruct: for instruction following and safer deployment
All variants are available in sizes of 7B, 13B and 34B parameters.
**This repository contains the base version of the 13B parameters model.**
**Input** Models input text only.
**Output** Models generate text only.
**Model Architecture** Code Llama is an auto-regressive language model that uses an optimized transformer architecture.
**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.
**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** 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/)" or its [arXiv page](https://arxiv.org/abs/2308.12950).
## 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).
## 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).
| 8,148 | [
[
-0.032196044921875,
-0.050750732421875,
0.0201873779296875,
0.041595458984375,
-0.013275146484375,
0.007778167724609375,
-0.0127410888671875,
-0.043304443359375,
0.0213623046875,
0.030029296875,
-0.030670166015625,
-0.04766845703125,
-0.042449951171875,
0.018096923828125,
-0.024566650390625,
0.0811767578125,
-0.0015516281127929688,
-0.02459716796875,
-0.00962066650390625,
-0.0022411346435546875,
-0.02105712890625,
-0.04168701171875,
-0.019500732421875,
-0.0298919677734375,
0.018707275390625,
0.0208740234375,
0.053924560546875,
0.0467529296875,
0.0404052734375,
0.027984619140625,
-0.021026611328125,
0.0041961669921875,
-0.0213623046875,
-0.0223846435546875,
0.011627197265625,
-0.0340576171875,
-0.0623779296875,
-0.0019130706787109375,
0.032073974609375,
0.0262298583984375,
-0.0172576904296875,
0.031707763671875,
-0.003635406494140625,
0.0423583984375,
-0.0241241455078125,
0.0242767333984375,
-0.043121337890625,
-0.0037479400634765625,
0.005100250244140625,
-0.0129852294921875,
-0.013458251953125,
-0.042083740234375,
-0.007007598876953125,
-0.036590576171875,
-0.0006237030029296875,
-0.00125885009765625,
0.08941650390625,
0.0297698974609375,
-0.0171051025390625,
-0.0191497802734375,
-0.023468017578125,
0.059173583984375,
-0.0704345703125,
0.00689697265625,
0.0245819091796875,
-0.00899505615234375,
-0.01493072509765625,
-0.06024169921875,
-0.052337646484375,
-0.0232696533203125,
-0.01088714599609375,
0.00743865966796875,
-0.029266357421875,
0.000820159912109375,
0.027862548828125,
0.03558349609375,
-0.040924072265625,
0.00998687744140625,
-0.031890869140625,
-0.0189666748046875,
0.067626953125,
0.01480865478515625,
0.0311431884765625,
-0.0260162353515625,
-0.026153564453125,
-0.0070953369140625,
-0.052764892578125,
0.01169586181640625,
0.030029296875,
-0.0037517547607421875,
-0.056488037109375,
0.04718017578125,
-0.014129638671875,
0.045654296875,
0.00406646728515625,
-0.035919189453125,
0.04644775390625,
-0.0274658203125,
-0.0211029052734375,
-0.013397216796875,
0.07452392578125,
0.04193115234375,
0.0190887451171875,
0.00696563720703125,
-0.012847900390625,
0.021484375,
0.0028553009033203125,
-0.06414794921875,
-0.012298583984375,
0.0291290283203125,
-0.044952392578125,
-0.04486083984375,
-0.013671875,
-0.06524658203125,
-0.0033435821533203125,
-0.0027866363525390625,
0.0186920166015625,
-0.021881103515625,
-0.034698486328125,
0.02130126953125,
0.001255035400390625,
0.0330810546875,
0.011474609375,
-0.06365966796875,
0.00708770751953125,
0.03741455078125,
0.0621337890625,
0.0018177032470703125,
-0.03466796875,
-0.0020122528076171875,
-0.00885009765625,
-0.02325439453125,
0.0418701171875,
-0.024383544921875,
-0.0311126708984375,
-0.01306915283203125,
0.01108551025390625,
-0.006000518798828125,
-0.03289794921875,
0.018707275390625,
-0.0227813720703125,
0.00395965576171875,
0.004180908203125,
-0.0294342041015625,
-0.028472900390625,
0.00289154052734375,
-0.039337158203125,
0.08258056640625,
0.016693115234375,
-0.05950927734375,
-0.0024852752685546875,
-0.048583984375,
-0.02239990234375,
-0.0181427001953125,
-0.0027446746826171875,
-0.054931640625,
-0.00772857666015625,
0.01995849609375,
0.03973388671875,
-0.0256500244140625,
0.0234375,
-0.01515960693359375,
-0.025543212890625,
0.015289306640625,
-0.0094146728515625,
0.080322265625,
0.0248565673828125,
-0.040008544921875,
0.013580322265625,
-0.059356689453125,
0.001613616943359375,
0.037689208984375,
-0.034881591796875,
0.01189422607421875,
-0.01251983642578125,
0.0029087066650390625,
0.00478363037109375,
0.03765869140625,
-0.0272979736328125,
0.034149169921875,
-0.030792236328125,
0.057830810546875,
0.053436279296875,
0.00041937828063964844,
0.02679443359375,
-0.04345703125,
0.050750732421875,
-0.005634307861328125,
0.0174407958984375,
-0.0201416015625,
-0.059112548828125,
-0.07159423828125,
-0.028533935546875,
0.0080718994140625,
0.049774169921875,
-0.03717041015625,
0.05474853515625,
-0.006855010986328125,
-0.05926513671875,
-0.0382080078125,
0.00989532470703125,
0.03363037109375,
0.028839111328125,
0.029022216796875,
-0.0159149169921875,
-0.0577392578125,
-0.05877685546875,
0.0094757080078125,
-0.03350830078125,
0.01318359375,
0.016876220703125,
0.06390380859375,
-0.037689208984375,
0.061859130859375,
-0.037994384765625,
-0.007442474365234375,
-0.02752685546875,
-0.019805908203125,
0.0467529296875,
0.0501708984375,
0.052520751953125,
-0.0474853515625,
-0.0237274169921875,
0.0000017881393432617188,
-0.0689697265625,
-0.007171630859375,
-0.0138702392578125,
-0.004573822021484375,
0.0294647216796875,
0.019683837890625,
-0.048004150390625,
0.040283203125,
0.05609130859375,
-0.0225372314453125,
0.053131103515625,
-0.013916015625,
-0.0006127357482910156,
-0.0797119140625,
0.0193023681640625,
-0.004726409912109375,
-0.0014829635620117188,
-0.0350341796875,
0.0250396728515625,
0.009552001953125,
0.007389068603515625,
-0.037017822265625,
0.03497314453125,
-0.0338134765625,
-0.0004773139953613281,
-0.00408935546875,
-0.00820159912109375,
-0.0003006458282470703,
0.055908203125,
0.0037784576416015625,
0.0654296875,
0.050384521484375,
-0.045257568359375,
0.0302734375,
0.024932861328125,
-0.0254974365234375,
0.0160369873046875,
-0.07000732421875,
0.015899658203125,
0.01204681396484375,
0.027862548828125,
-0.0657958984375,
-0.0151214599609375,
0.024444580078125,
-0.045135498046875,
0.008819580078125,
-0.00433349609375,
-0.037811279296875,
-0.044677734375,
-0.024993896484375,
0.028564453125,
0.06463623046875,
-0.044189453125,
0.0264739990234375,
0.031219482421875,
0.00865936279296875,
-0.046173095703125,
-0.05059814453125,
0.0026607513427734375,
-0.032989501953125,
-0.058502197265625,
0.030029296875,
-0.0217742919921875,
-0.006000518798828125,
-0.013763427734375,
0.003864288330078125,
0.0013532638549804688,
0.0186767578125,
0.0345458984375,
0.03741455078125,
-0.0148468017578125,
-0.0172576904296875,
-0.01073455810546875,
-0.01412200927734375,
0.004669189453125,
-0.001117706298828125,
0.054595947265625,
-0.03448486328125,
-0.016204833984375,
-0.0499267578125,
0.0082550048828125,
0.045684814453125,
-0.021392822265625,
0.05078125,
0.037078857421875,
-0.02447509765625,
-0.0006012916564941406,
-0.04815673828125,
-0.0008840560913085938,
-0.04205322265625,
0.0227508544921875,
-0.0201873779296875,
-0.05963134765625,
0.05279541015625,
0.01284027099609375,
0.01126861572265625,
0.038177490234375,
0.059906005859375,
0.0055084228515625,
0.0604248046875,
0.06048583984375,
-0.0289154052734375,
0.0264434814453125,
-0.04998779296875,
0.006988525390625,
-0.06048583984375,
-0.0335693359375,
-0.04425048828125,
-0.00632476806640625,
-0.052978515625,
-0.0302581787109375,
0.0292205810546875,
0.01387786865234375,
-0.03631591796875,
0.054473876953125,
-0.06585693359375,
0.0247802734375,
0.03363037109375,
0.00862884521484375,
0.0233001708984375,
0.005474090576171875,
-0.01073455810546875,
0.0125885009765625,
-0.0390625,
-0.0491943359375,
0.08660888671875,
0.03619384765625,
0.06390380859375,
-0.002170562744140625,
0.060150146484375,
0.003322601318359375,
0.0193023681640625,
-0.04315185546875,
0.045379638671875,
0.0213470458984375,
-0.037078857421875,
-0.004512786865234375,
-0.025726318359375,
-0.0693359375,
0.0160064697265625,
0.0028705596923828125,
-0.0667724609375,
0.00604248046875,
0.003582000732421875,
-0.0220947265625,
0.0295257568359375,
-0.05169677734375,
0.056976318359375,
-0.00873565673828125,
-0.0081024169921875,
-0.007232666015625,
-0.044189453125,
0.03546142578125,
0.0017633438110351562,
0.01715087890625,
-0.008819580078125,
-0.0133819580078125,
0.054779052734375,
-0.0467529296875,
0.07257080078125,
0.004058837890625,
-0.0262298583984375,
0.0399169921875,
-0.0024738311767578125,
0.032806396484375,
0.005401611328125,
-0.0195770263671875,
0.044769287109375,
-0.00357818603515625,
-0.0248260498046875,
-0.008819580078125,
0.051849365234375,
-0.07684326171875,
-0.053192138671875,
-0.035980224609375,
-0.025970458984375,
0.01477813720703125,
0.00865936279296875,
0.0341796875,
0.014068603515625,
0.014495849609375,
0.0120849609375,
0.0341796875,
-0.046661376953125,
0.052276611328125,
0.03216552734375,
-0.020294189453125,
-0.04193115234375,
0.066162109375,
-0.006931304931640625,
0.01690673828125,
0.017059326171875,
0.0035839080810546875,
-0.0160369873046875,
-0.03363037109375,
-0.0341796875,
0.0335693359375,
-0.039947509765625,
-0.04132080078125,
-0.05426025390625,
-0.0345458984375,
-0.031402587890625,
-0.021331787109375,
-0.0245819091796875,
-0.023223876953125,
-0.044158935546875,
-0.0109405517578125,
0.056243896484375,
0.050262451171875,
-0.0084075927734375,
0.022247314453125,
-0.048553466796875,
0.0302581787109375,
0.00727081298828125,
0.028045654296875,
0.006237030029296875,
-0.04290771484375,
-0.01087188720703125,
-0.0007386207580566406,
-0.044647216796875,
-0.06768798828125,
0.04632568359375,
0.00844573974609375,
0.04193115234375,
0.004688262939453125,
-0.0017633438110351562,
0.052154541015625,
-0.02752685546875,
0.068115234375,
0.02838134765625,
-0.08502197265625,
0.050872802734375,
-0.0212554931640625,
0.00984954833984375,
0.0043792724609375,
0.0225677490234375,
-0.0345458984375,
-0.0290679931640625,
-0.053741455078125,
-0.062225341796875,
0.052276611328125,
0.02362060546875,
0.0115509033203125,
0.0033245086669921875,
0.0251617431640625,
-0.006443023681640625,
0.0162506103515625,
-0.0777587890625,
-0.032501220703125,
-0.026123046875,
-0.01451873779296875,
-0.0036029815673828125,
-0.00914764404296875,
0.0031871795654296875,
-0.02783203125,
0.04150390625,
-0.0087738037109375,
0.043975830078125,
0.0197296142578125,
-0.01209259033203125,
-0.0168304443359375,
0.002231597900390625,
0.0487060546875,
0.04669189453125,
-0.004497528076171875,
-0.01152801513671875,
0.0258636474609375,
-0.0413818359375,
0.0187530517578125,
-0.007167816162109375,
-0.00714111328125,
-0.01338958740234375,
0.04150390625,
0.047882080078125,
0.003841400146484375,
-0.047149658203125,
0.04193115234375,
0.006412506103515625,
-0.0185699462890625,
-0.033416748046875,
0.0179443359375,
0.0274505615234375,
0.020904541015625,
0.0229949951171875,
-0.0002092123031616211,
-0.01031494140625,
-0.0293731689453125,
0.00753021240234375,
0.022491455078125,
0.006771087646484375,
-0.02728271484375,
0.06878662109375,
0.007320404052734375,
-0.0189361572265625,
0.033111572265625,
-0.001209259033203125,
-0.050506591796875,
0.09039306640625,
0.04669189453125,
0.0545654296875,
-0.0174560546875,
0.01239013671875,
0.044403076171875,
0.037994384765625,
0.000614166259765625,
0.031402587890625,
0.004993438720703125,
-0.03912353515625,
-0.0189971923828125,
-0.06341552734375,
-0.0173797607421875,
0.00778961181640625,
-0.037139892578125,
0.0259246826171875,
-0.048065185546875,
-0.010284423828125,
-0.0160369873046875,
0.015716552734375,
-0.0487060546875,
0.003505706787109375,
0.0035152435302734375,
0.07220458984375,
-0.053375244140625,
0.0654296875,
0.04327392578125,
-0.052978515625,
-0.07354736328125,
-0.017852783203125,
-0.0018825531005859375,
-0.0810546875,
0.041259765625,
0.0172882080078125,
0.00905609130859375,
0.00321197509765625,
-0.06182861328125,
-0.0748291015625,
0.093017578125,
0.0297393798828125,
-0.03436279296875,
-0.0006976127624511719,
0.00897216796875,
0.043975830078125,
-0.027496337890625,
0.039215087890625,
0.04290771484375,
0.0352783203125,
-0.008453369140625,
-0.09063720703125,
0.0256500244140625,
-0.030670166015625,
0.01465606689453125,
-0.0152587890625,
-0.0831298828125,
0.07940673828125,
-0.038421630859375,
-0.00478363037109375,
0.030029296875,
0.05535888671875,
0.04156494140625,
0.00980377197265625,
0.02166748046875,
0.045745849609375,
0.04156494140625,
-0.0068359375,
0.0791015625,
-0.03363037109375,
0.046142578125,
0.044189453125,
-0.00373077392578125,
0.0531005859375,
0.022430419921875,
-0.038909912109375,
0.053558349609375,
0.058135986328125,
-0.0176239013671875,
0.0237884521484375,
0.023590087890625,
-0.00380706787109375,
-0.004871368408203125,
0.0025787353515625,
-0.06304931640625,
0.028778076171875,
0.0245208740234375,
-0.0261993408203125,
0.0013933181762695312,
-0.01303863525390625,
0.0233917236328125,
-0.0104522705078125,
-0.005924224853515625,
0.0489501953125,
0.01471710205078125,
-0.04168701171875,
0.0897216796875,
0.00662994384765625,
0.0726318359375,
-0.035888671875,
-0.00579071044921875,
-0.031402587890625,
0.00809478759765625,
-0.04229736328125,
-0.047515869140625,
0.00957489013671875,
0.0147857666015625,
-0.001705169677734375,
-0.0087738037109375,
0.0306854248046875,
-0.01264190673828125,
-0.0379638671875,
0.030792236328125,
0.01338958740234375,
0.0214385986328125,
0.01267242431640625,
-0.061370849609375,
0.0335693359375,
0.01512908935546875,
-0.034515380859375,
0.021453857421875,
0.015106201171875,
0.01255035400390625,
0.0665283203125,
0.05267333984375,
-0.0015459060668945312,
0.0134735107421875,
-0.016937255859375,
0.080322265625,
-0.055450439453125,
-0.03350830078125,
-0.0654296875,
0.046142578125,
0.0120697021484375,
-0.033203125,
0.05206298828125,
0.03765869140625,
0.058380126953125,
-0.01126861572265625,
0.057037353515625,
-0.0226898193359375,
0.01000213623046875,
-0.034820556640625,
0.0596923828125,
-0.05487060546875,
0.0235595703125,
-0.043212890625,
-0.0684814453125,
-0.0186309814453125,
0.06494140625,
-0.004150390625,
0.00894927978515625,
0.046966552734375,
0.0811767578125,
0.0156707763671875,
-0.0017681121826171875,
0.01126861572265625,
0.0226898193359375,
0.02825927734375,
0.069091796875,
0.06378173828125,
-0.049041748046875,
0.04730224609375,
-0.044189453125,
-0.022430419921875,
-0.0246429443359375,
-0.07183837890625,
-0.0672607421875,
-0.04449462890625,
-0.032806396484375,
-0.037933349609375,
-0.0136260986328125,
0.07427978515625,
0.052093505859375,
-0.051055908203125,
-0.037506103515625,
-0.00528717041015625,
0.0298614501953125,
-0.0126953125,
-0.0170135498046875,
0.03143310546875,
-0.012451171875,
-0.0596923828125,
0.01739501953125,
0.004497528076171875,
0.00847625732421875,
-0.01824951171875,
-0.0227813720703125,
-0.0104827880859375,
-0.0012960433959960938,
0.03045654296875,
0.0308837890625,
-0.06072998046875,
-0.01666259765625,
0.0008516311645507812,
-0.0221405029296875,
0.0159759521484375,
0.0272674560546875,
-0.047149658203125,
0.0005764961242675781,
0.032501220703125,
0.02655029296875,
0.029022216796875,
-0.0158538818359375,
0.0158843994140625,
-0.0258941650390625,
0.02691650390625,
-0.0005846023559570312,
0.0377197265625,
0.0105438232421875,
-0.038055419921875,
0.0523681640625,
0.02691650390625,
-0.049774169921875,
-0.066650390625,
0.0018587112426757812,
-0.0843505859375,
-0.01338958740234375,
0.0933837890625,
-0.017547607421875,
-0.0291748046875,
0.01134490966796875,
-0.0256500244140625,
0.0190887451171875,
-0.03662109375,
0.051544189453125,
0.0226287841796875,
-0.006763458251953125,
-0.00435638427734375,
-0.0297698974609375,
0.0246429443359375,
0.0245361328125,
-0.07220458984375,
-0.0120697021484375,
0.02569580078125,
0.03338623046875,
0.0166168212890625,
0.056304931640625,
-0.01256561279296875,
0.0170440673828125,
-0.00025963783264160156,
0.0304412841796875,
-0.007328033447265625,
-0.012237548828125,
-0.0246429443359375,
0.002483367919921875,
-0.01221466064453125,
-0.0099334716796875
]
] |
cerebras/Cerebras-GPT-2.7B | 2023-04-07T13:51:01.000Z | [
"transformers",
"pytorch",
"gpt2",
"causal-lm",
"text-generation",
"en",
"dataset:the_pile",
"arxiv:2304.03208",
"arxiv:2203.15556",
"arxiv:2101.00027",
"license:apache-2.0",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | cerebras | null | null | cerebras/Cerebras-GPT-2.7B | 41 | 7,466 | transformers | 2023-03-20T20:44:46 | ---
language:
- en
inference: false
tags:
- pytorch
- causal-lm
license: apache-2.0
datasets:
- the_pile
pipeline_tag: text-generation
---
# Cerebras-GPT 2.7B
Check out our [Blog Post](https://www.cerebras.net/cerebras-gpt) and [arXiv paper](https://arxiv.org/abs/2304.03208)!
## Model Description
The Cerebras-GPT family is released to facilitate research into LLM scaling laws using open architectures and data sets and demonstrate the simplicity of and scalability of training LLMs on the Cerebras software and hardware stack. All Cerebras-GPT models are available on Hugging Face.
The family includes 111M, 256M, 590M, 1.3B, 2.7B, 6.7B, and 13B models.
All models in the Cerebras-GPT family have been trained in accordance with [Chinchilla scaling laws](https://arxiv.org/abs/2203.15556) (20 tokens per model parameter) which is compute-optimal.
These models were trained on the [Andromeda](https://www.cerebras.net/andromeda/) AI supercomputer comprised of 16 CS-2 wafer scale systems. Cerebras' [weight streaming technology](https://www.cerebras.net/blog/linear-scaling-made-possible-with-weight-streaming) simplifies the training of LLMs by disaggregating compute from model storage. This allowed for efficient scaling of training across nodes using simple data parallelism.
Cerebras systems for pre-training and fine tuning are available in the cloud via the [Cerebras Model Studio](https://www.cerebras.net/product-cloud/). Cerebras CS-2 compatible checkpoints are available in [Cerebras Model Zoo](https://github.com/Cerebras/modelzoo).
## Model Details
* Developed by: [Cerebras Systems](https://www.cerebras.net/)
* License: Apache 2.0
* Model type: Transformer-based Language Model
* Architecture: GPT-3 style architecture
* Data set: The Pile
* Tokenizer: Byte Pair Encoding
* Vocabulary Size: 50257
* Sequence Length: 2048
* Optimizer: AdamW, (β1, β2) = (0.9, 0.95), adam_eps = 1e−8 (1e−9 for larger models)
* Positional Encoding: Learned
* Language: English
* Learn more: Dense Scaling Laws Paper for training procedure, config files, and details on how to use.
**Contact**: To ask questions about Cerebras-GPT models, join the [Cerebras Discord](https://discord.gg/q6bZcMWJVu).
This is the standard parameterization version of Cerebras-GPT with **2.7B** parameters
Related models: [Cerebras-GPT Models](https://huggingface.co/models?sort=downloads&search=cerebras-gpt)
<br><br>
| Model | Parameters | Layers | d_model | Heads | d_head | d_ffn | LR | BS (seq) | BS (tokens) |
|---------------|------------|--------|---------|-------|--------|--------|----------|----------|----------------|
| Cerebras-GPT | 111M | 10 | 768 | 12 | 64 | 3072 | 6.0E-04 | 120 | 246K |
| Cerebras-GPT | 256M | 14 | 1088 | 17 | 64 | 4352 | 6.0E-04 | 264 | 541K |
| Cerebras-GPT | 590M | 18 | 1536 | 12 | 128 | 6144 | 2.0E-04 | 264 | 541K |
| Cerebras-GPT | 1.3B | 24 | 2048 | 16 | 128 | 8192 | 2.0E-04 | 528 | 1.08M |
| Cerebras-GPT | 2.7B | 32 | 2560 | 20 | 128 | 10240 | 2.0E-04 | 528 | 1.08M |
| Cerebras-GPT | 6.7B | 32 | 4096 | 32 | 128 | 16384 | 1.2E-04 | 1040 | 2.13M |
| Cerebras-GPT | 13B | 40 | 5120 | 40 | 128 | 20480 | 1.2E-04 | 720 → 1080 | 1.47M → 2.21M |
<br><br>
## Quickstart
This model can be easily loaded using the AutoModelForCausalLM functionality:
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("cerebras/Cerebras-GPT-2.7B")
model = AutoModelForCausalLM.from_pretrained("cerebras/Cerebras-GPT-2.7B")
text = "Generative AI is "
```
And can be used with Hugging Face Pipelines
```python
from transformers import pipeline
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
generated_text = pipe(text, max_length=50, do_sample=False, no_repeat_ngram_size=2)[0]
print(generated_text['generated_text'])
```
or with `model.generate()`
```python
inputs = tokenizer(text, return_tensors="pt")
outputs = model.generate(**inputs, num_beams=5,
max_new_tokens=50, early_stopping=True,
no_repeat_ngram_size=2)
text_output = tokenizer.batch_decode(outputs, skip_special_tokens=True)
print(text_output[0])
```
<br><br>
## Training data
Cerebras-GPT is trained using [the Pile](https://pile.eleuther.ai) dataset from [EleutherAI](https://www.eleuther.ai). See the [Pile paper](https://arxiv.org/abs/2101.00027) for a more detailed breakdown of data sources and methodology. The Pile was cleaned using the ftfy library to normalize the text, then filtered using scripts provided by Eleuther.
We tokenized the data using byte-pair encoding using the GPT-2 vocabulary. Our tokenized version of the Pile has 371B tokens. We include more details about the training dataset preprocessing in Appendix A.1 of our paper.
Recent works find significant duplicate data present in the Pile. Eleuther’s Pythia applies a deduplication process to reduce replicated data, decreasing the Pile dataset size. Pythia was trained on both the standard dataset and deduplicated dataset to characterize the impact. Our models are trained on the standard Pile without deduplication, which may present an opportunity for further improvement with the deduplicated data set.
<br><br>
## Training procedure
We use the GPT-3 style model architecture. All of our layers use full attention as opposed to the GPT-3 style sparse banded attention. The model shapes were selected to either follow aspect ratio 80 or are the same shape as GPT-3 models. Learning rate warmed up for 375M tokens (1500 steps for 111M and 256M models) and 10x cosine decayed. No dropout was used and weight decay was set to 0.1. All models are trained with MSL of 2048.
All models were trained to Chinchilla point: 20 tokens per model parameter. Number of steps was chosen based on optimal batch size (varied by model) and fixed sequence length (2048).See Training Table, below, for details.
<br>
Model Params | Sequence Length | Batch Size | Number of Steps | Tokens | Tokens per Parameter | Flops
------------ | -------------- | ---------- | --------------- | ------ | -------------------- | -----
111M | 2048 | 120 | 9037 | 2.22E+09 | 20 | 2.6E+18
256M | 2048 | 264 | 9468 | 5.12E+09 | 20 | 1.3E+19
590M | 2048 | 264 | 21836 | 1.18E+10 | 20 | 6.1E+19
1.3B | 2048 | 528 | 24334 | 2.63E+10 | 20 | 2.8E+20
2.7B | 2048 | 528 | 49041 | 5.30E+10 | 20 | 1.1E+21
6.7B | 2048 | 1040 | 62522 | 1.33E+11 | 20 | 6.3E+21
13B | 2048 | 720 | 174335 | 2.57E+11 | 20 | 2.3E+22
<br><br>
## Evaluations
We trained models from smallest to largest and fit a power law as we went along. The power law was helpful for extrapolating the validation loss of the next largest model we trained and provided confidence about whether the training run was going well.
We performed upstream (pre-training) evaluations of text prediction cross-entropy using the Pile validation and test splits. We performed downstream evaluations of text generation accuracy on standardized tasks using the [Eleuther lm-evaluation-harness](https://github.com/EleutherAI/lm-evaluation-harness). Results are compared against many publicly available large language models in Section 3 of the paper.
#### 0-shot Evaluation
| Model | Params | Training FLOPs | PILE test xent | Hella-Swag | PIQA | Wino-Grande | Lambada | ARC-e | ARC-c | OpenBookQA | Downstream Average |
| ------- | ----- | -------------- | -------------- | ---------- | ----- | ----------- | ------- | ----- | ----- | ---------- | ------------------ |
| Cerebras-GPT | 111M | 2.6E+18 | 2.566 | 0.268 | 0.594 | 0.488 | 0.194 | 0.380 | 0.166 | 0.118 | 0.315 |
| Cerebras-GPT | 256M | 1.3E+19 | 2.299 | 0.274 | 0.613 | 0.511 | 0.293 | 0.410 | 0.170 | 0.158 | 0.347 |
| Cerebras-GPT | 590M | 6.1E+19 | 2.184 | 0.291 | 0.627 | 0.498 | 0.366 | 0.464 | 0.190 | 0.158 | 0.370 |
| Cerebras-GPT | 1.3B | 2.8E+20 | 1.996 | 0.325 | 0.664 | 0.521 | 0.462 | 0.508 | 0.224 | 0.166 | 0.410 |
| Cerebras-GPT | 2.7B | 1.1E+21 | 1.834 | 0.386 | 0.701 | 0.559 | 0.567 | 0.571 | 0.246 | 0.206 | 0.462 |
| Cerebras-GPT | 6.7B | 6.3E+21 | 1.704 | 0.447 | 0.739 | 0.602 | 0.636 | 0.643 | 0.282 | 0.238 | 0.512 |
| Cerebras-GPT | 13B | 2.3E+22 | 1.575 | 0.513 | 0.766 | 0.646 | 0.696 | 0.714 | 0.367 | 0.286 | 0.570 |
#### 5-shot Evaluation
| Model | Params | Hella-Swag | PIQA | Wino-Grande | Lambada | ARC-e | ARC-c | OpenBookQA |
| -------- | ----- | ----------| ----- | ----------- | -------| ----- | ----- | ---------- |
| Cerebras-GPT | 111M | 0.267 | 0.588 | 0.475 | 0.158 | 0.356 | 0.166 | 0.136 |
| Cerebras-GPT | 256M | 0.278 | 0.606 | 0.522 | 0.225 | 0.422 | 0.183 | 0.164 |
| Cerebras-GPT | 590M | 0.291 | 0.634 | 0.479 | 0.281 | 0.475 | 0.206 | 0.152 |
| Cerebras-GPT | 1.3B | 0.326 | 0.668 | 0.536 | 0.395 | 0.529 | 0.241 | 0.174 |
| Cerebras-GPT | 2.7B | 0.382 | 0.697 | 0.543 | 0.487 | 0.590 | 0.267 | 0.224 |
| Cerebras-GPT | 6.7B | 0.444 | 0.736 | 0.590 | 0.591 | 0.667 | 0.314 | 0.270 |
| Cerebras-GPT | 13B | 0.514 | 0.768 | 0.674 | 0.655 | 0.743 | 0.398 | 0.318 |
<br><br>
## Uses and Limitations
### Intended Use
The primary intended use is to further research into large language models. These models can be used as a foundation model for NLP, applications, ethics, and alignment research. Our primary intended users are researchers who are working to improve LLMs and practitioners seeking reference implementations, training setups, hyperparameters, or pre-trained models. We release these models with a fully permissive Apache license for the community to use freely.
You may fine-tune and adapt Cerebras-GPT models for deployment via either Cerebras [Model Studio](https://www.cerebras.net/product-cloud/) or third-party libraries. Further safety-related testing and mitigations should be applied beore using the Cerebras-GPT model family in production downstream applications.
Due to financial and compute budgets, Cerebras-GPT models were only trained and evaluated following the approaches described in the paper.
### Out of Scope Use
Cerebras-GPT models are trained on the Pile, with English language only, and are not suitable for machine translation tasks.
Cerebras-GPT models have not been tuned for human-facing dialog applications like chatbots and will not respond to prompts in a similar way to models that have received instruction tuning or reinforcement learning from human feedback (RLHF) like Flan-T5 or ChatGPT. Cerebras-GPT models can be tuned using those methods.
### Risk, Bias, Ethical Considerations
* **Data**: The Pile dataset has been thoroughly analyzed from various ethical standpoints such as toxicity analysis, gender bias, pejorative content, racially sensitive content etc. Please refer to Pile dataset references.
* **Human life**: The outputs from this model may or may not align with human values. The risk needs to be thoroughly investigated before deploying this model in a production environment where it can directly impact human life.
* **Risks and harms**: There can be distributional bias in the Pile dataset that can manifest in various forms in the downstream model deployment. There are other risks associated with large language models such as amplifying stereotypes, memorizing training data, or revealing private or secure information.
* **Mitigations**: Only mitigations in standard Pile dataset pre-processing were employed when pre-training Cerebras-GPT.
<br><br>
## Acknowledgements
We are thankful to all Cerebras engineers, past and present, that made this work possible. | 12,575 | [
[
-0.0272369384765625,
-0.04766845703125,
0.0181884765625,
0.0133056640625,
-0.0196533203125,
-0.01520538330078125,
-0.0160675048828125,
-0.03253173828125,
0.01270294189453125,
0.0215301513671875,
-0.028045654296875,
-0.03033447265625,
-0.05621337890625,
-0.0165557861328125,
-0.03021240234375,
0.0855712890625,
-0.006465911865234375,
0.003345489501953125,
0.0104522705078125,
-0.005405426025390625,
-0.0151824951171875,
-0.04193115234375,
-0.059356689453125,
-0.0304107666015625,
0.03521728515625,
-0.0006842613220214844,
0.05670166015625,
0.058929443359375,
0.0262603759765625,
0.021514892578125,
-0.0287322998046875,
-0.00423431396484375,
-0.0241241455078125,
-0.023529052734375,
0.010009765625,
-0.018707275390625,
-0.041107177734375,
-0.007076263427734375,
0.05126953125,
0.0478515625,
-0.0269622802734375,
0.018829345703125,
0.025787353515625,
0.054901123046875,
-0.036285400390625,
0.0132598876953125,
-0.03717041015625,
0.00106048583984375,
-0.018890380859375,
0.0010509490966796875,
-0.021728515625,
-0.0149383544921875,
0.0018310546875,
-0.040130615234375,
0.0217742919921875,
-0.0036602020263671875,
0.09552001953125,
0.017303466796875,
-0.03277587890625,
-0.0205230712890625,
-0.03143310546875,
0.052764892578125,
-0.05657958984375,
0.0290069580078125,
0.01361846923828125,
-0.00039267539978027344,
-0.00244903564453125,
-0.064697265625,
-0.03900146484375,
-0.016387939453125,
-0.015777587890625,
0.011871337890625,
-0.0162200927734375,
0.0038127899169921875,
0.033782958984375,
0.0386962890625,
-0.059967041015625,
0.0160675048828125,
-0.037109375,
-0.0181427001953125,
0.05096435546875,
0.01080322265625,
0.01513671875,
-0.0259857177734375,
-0.031951904296875,
-0.03009033203125,
-0.0380859375,
0.025054931640625,
0.030792236328125,
0.01494598388671875,
-0.031890869140625,
0.0298004150390625,
-0.01227569580078125,
0.045318603515625,
0.02215576171875,
-0.00684356689453125,
0.040924072265625,
-0.021484375,
-0.03277587890625,
-0.005260467529296875,
0.07818603515625,
0.0136260986328125,
0.01248931884765625,
0.007335662841796875,
-0.01479339599609375,
-0.011138916015625,
-0.00022709369659423828,
-0.0816650390625,
-0.0273284912109375,
0.01337432861328125,
-0.0438232421875,
-0.0296630859375,
0.0027217864990234375,
-0.052703857421875,
-0.0154266357421875,
-0.03033447265625,
0.036590576171875,
-0.03643798828125,
-0.026153564453125,
0.00604248046875,
0.0019817352294921875,
0.033447265625,
0.019287109375,
-0.089111328125,
0.0211944580078125,
0.03082275390625,
0.06378173828125,
0.0036792755126953125,
-0.0279693603515625,
-0.0180816650390625,
-0.0013799667358398438,
-0.01161956787109375,
0.03643798828125,
-0.0023136138916015625,
-0.0269012451171875,
-0.016815185546875,
0.009796142578125,
-0.032012939453125,
-0.02703857421875,
0.0377197265625,
-0.0263214111328125,
0.016754150390625,
-0.011688232421875,
-0.03955078125,
-0.028106689453125,
0.01181793212890625,
-0.040924072265625,
0.08477783203125,
0.0154266357421875,
-0.0692138671875,
0.0198974609375,
-0.034454345703125,
-0.0189208984375,
-0.006191253662109375,
-0.01059722900390625,
-0.04791259765625,
-0.01187896728515625,
0.03192138671875,
0.04217529296875,
-0.0250244140625,
0.0259552001953125,
-0.01715087890625,
-0.02105712890625,
-0.00677490234375,
-0.0377197265625,
0.0877685546875,
0.0216522216796875,
-0.04656982421875,
-0.0007419586181640625,
-0.054168701171875,
0.0106658935546875,
0.0268402099609375,
-0.03021240234375,
0.00821685791015625,
-0.0160675048828125,
0.0089874267578125,
0.0196533203125,
0.0274200439453125,
-0.0196533203125,
0.0143890380859375,
-0.03387451171875,
0.039764404296875,
0.05303955078125,
0.003345489501953125,
0.02276611328125,
-0.0227203369140625,
0.033233642578125,
0.006244659423828125,
0.0179290771484375,
-0.00981903076171875,
-0.039703369140625,
-0.05645751953125,
-0.0175933837890625,
0.0330810546875,
0.0419921875,
-0.033660888671875,
0.037017822265625,
-0.0222625732421875,
-0.05938720703125,
-0.0158843994140625,
0.005680084228515625,
0.03607177734375,
0.040313720703125,
0.03277587890625,
-0.0198822021484375,
-0.036590576171875,
-0.07196044921875,
-0.00579071044921875,
-0.0191650390625,
-0.003826141357421875,
0.01548004150390625,
0.057647705078125,
-0.003936767578125,
0.053558349609375,
-0.035430908203125,
-0.0047149658203125,
-0.005767822265625,
0.01447296142578125,
0.0323486328125,
0.047149658203125,
0.045928955078125,
-0.0572509765625,
-0.04119873046875,
0.001438140869140625,
-0.0609130859375,
0.008697509765625,
-0.0151214599609375,
0.0033130645751953125,
0.02288818359375,
0.033111572265625,
-0.05474853515625,
0.0274200439453125,
0.04833984375,
-0.025634765625,
0.047698974609375,
-0.0202484130859375,
-0.0011758804321289062,
-0.0804443359375,
0.02276611328125,
0.0107421875,
-0.00287628173828125,
-0.04412841796875,
0.00533294677734375,
0.018402099609375,
0.0032329559326171875,
-0.045318603515625,
0.04052734375,
-0.045440673828125,
-0.0007357597351074219,
-0.0004792213439941406,
0.00998687744140625,
-0.0066680908203125,
0.06353759765625,
0.0076141357421875,
0.051910400390625,
0.04638671875,
-0.047271728515625,
0.0090484619140625,
0.01210784912109375,
-0.0164337158203125,
0.0267333984375,
-0.0628662109375,
0.0029315948486328125,
-0.00225830078125,
0.0272369384765625,
-0.0543212890625,
-0.01329803466796875,
0.0190582275390625,
-0.044921875,
0.038543701171875,
-0.0205535888671875,
-0.03155517578125,
-0.04852294921875,
-0.0225830078125,
0.024627685546875,
0.053131103515625,
-0.043731689453125,
0.041259765625,
0.0200653076171875,
-0.0034942626953125,
-0.04925537109375,
-0.05474853515625,
-0.0028934478759765625,
-0.0290985107421875,
-0.06378173828125,
0.038482666015625,
-0.00492095947265625,
-0.00029850006103515625,
-0.01483154296875,
0.0034046173095703125,
0.0028324127197265625,
0.002140045166015625,
0.022857666015625,
0.0210113525390625,
-0.01016998291015625,
-0.007762908935546875,
0.002628326416015625,
-0.006748199462890625,
0.006237030029296875,
-0.026519775390625,
0.052734375,
-0.0291748046875,
-0.017364501953125,
-0.039642333984375,
-0.011932373046875,
0.0428466796875,
-0.0140380859375,
0.0628662109375,
0.061798095703125,
-0.04022216796875,
0.01268768310546875,
-0.03509521484375,
-0.002288818359375,
-0.037322998046875,
0.037445068359375,
-0.02947998046875,
-0.05413818359375,
0.054779052734375,
0.022003173828125,
0.00788116455078125,
0.06304931640625,
0.056854248046875,
0.00792694091796875,
0.083740234375,
0.0283355712890625,
-0.0137176513671875,
0.036651611328125,
-0.052642822265625,
0.00039267539978027344,
-0.0723876953125,
-0.020843505859375,
-0.03253173828125,
-0.013427734375,
-0.053009033203125,
-0.022430419921875,
0.0185699462890625,
0.028076171875,
-0.050079345703125,
0.037841796875,
-0.055511474609375,
0.0166015625,
0.035614013671875,
0.01416778564453125,
0.005054473876953125,
0.0014066696166992188,
-0.0236663818359375,
0.00015282630920410156,
-0.052581787109375,
-0.03607177734375,
0.0924072265625,
0.04193115234375,
0.0328369140625,
-0.007625579833984375,
0.057403564453125,
-0.0013742446899414062,
0.02923583984375,
-0.04620361328125,
0.032958984375,
-0.006473541259765625,
-0.046783447265625,
-0.0242462158203125,
-0.042816162109375,
-0.07525634765625,
0.0382080078125,
0.002017974853515625,
-0.07403564453125,
0.0198822021484375,
0.00914764404296875,
-0.033538818359375,
0.0443115234375,
-0.043304443359375,
0.0699462890625,
-0.0194549560546875,
-0.0289459228515625,
-0.011474609375,
-0.05487060546875,
0.035858154296875,
-0.0022735595703125,
0.0171051025390625,
0.01039886474609375,
0.00576019287109375,
0.07171630859375,
-0.05084228515625,
0.0537109375,
-0.0255126953125,
-0.012359619140625,
0.042205810546875,
-0.00867462158203125,
0.05902099609375,
0.00008291006088256836,
-0.00494384765625,
0.019012451171875,
0.0006489753723144531,
-0.0307464599609375,
-0.0190887451171875,
0.057464599609375,
-0.0804443359375,
-0.03472900390625,
-0.03936767578125,
-0.037384033203125,
0.00513458251953125,
0.01165008544921875,
0.038604736328125,
0.029296875,
0.002620697021484375,
0.027618408203125,
0.047760009765625,
-0.0132293701171875,
0.050933837890625,
0.0221099853515625,
-0.0164642333984375,
-0.046173095703125,
0.0616455078125,
0.0223236083984375,
0.0184783935546875,
0.01412200927734375,
0.00766754150390625,
-0.0306549072265625,
-0.04620361328125,
-0.04248046875,
0.02374267578125,
-0.047332763671875,
-0.00981903076171875,
-0.0604248046875,
-0.03125,
-0.035308837890625,
-0.00936126708984375,
-0.0249176025390625,
-0.030853271484375,
-0.025054931640625,
-0.006610870361328125,
0.026214599609375,
0.0389404296875,
-0.0083770751953125,
0.0283966064453125,
-0.054656982421875,
0.006473541259765625,
0.0238189697265625,
0.009429931640625,
0.01496124267578125,
-0.07305908203125,
-0.0260772705078125,
0.0089874267578125,
-0.04833984375,
-0.061492919921875,
0.043731689453125,
-0.004302978515625,
0.03387451171875,
0.0234222412109375,
-0.0220794677734375,
0.053619384765625,
-0.021575927734375,
0.07275390625,
0.0253448486328125,
-0.07135009765625,
0.03887939453125,
-0.04534912109375,
0.01448822021484375,
0.032257080078125,
0.0273284912109375,
-0.037994384765625,
-0.0135498046875,
-0.07318115234375,
-0.07403564453125,
0.056304931640625,
0.02679443359375,
-0.0009126663208007812,
0.011322021484375,
0.03533935546875,
-0.013519287109375,
0.0113372802734375,
-0.07720947265625,
-0.0211181640625,
-0.0222930908203125,
-0.0143585205078125,
-0.0022983551025390625,
0.0012874603271484375,
0.010711669921875,
-0.035614013671875,
0.065185546875,
-0.0078277587890625,
0.01837158203125,
0.0195465087890625,
-0.01374053955078125,
-0.01050567626953125,
-0.00394439697265625,
0.03955078125,
0.042083740234375,
-0.01136016845703125,
-0.019683837890625,
0.033050537109375,
-0.05621337890625,
0.00299835205078125,
0.0231475830078125,
-0.02685546875,
-0.00948333740234375,
0.0185089111328125,
0.07012939453125,
0.01352691650390625,
-0.02313232421875,
0.03497314453125,
0.0030498504638671875,
-0.042388916015625,
-0.03045654296875,
0.0001614093780517578,
0.01568603515625,
0.01544952392578125,
0.0275421142578125,
-0.0005955696105957031,
0.0019102096557617188,
-0.0220184326171875,
0.01026153564453125,
0.0267486572265625,
-0.0225067138671875,
-0.020355224609375,
0.07196044921875,
-0.00217437744140625,
-0.00705718994140625,
0.05084228515625,
-0.01451873779296875,
-0.0372314453125,
0.0755615234375,
0.024932861328125,
0.062744140625,
-0.020751953125,
0.01055145263671875,
0.061981201171875,
0.028839111328125,
-0.020660400390625,
0.0037841796875,
0.0069580078125,
-0.03619384765625,
-0.0211944580078125,
-0.060089111328125,
-0.01503753662109375,
0.0247344970703125,
-0.053497314453125,
0.036651611328125,
-0.03753662109375,
-0.00849151611328125,
-0.007030487060546875,
0.0249786376953125,
-0.05621337890625,
0.029510498046875,
0.0218048095703125,
0.06390380859375,
-0.063232421875,
0.06866455078125,
0.03857421875,
-0.054656982421875,
-0.0897216796875,
-0.0052032470703125,
-0.0014123916625976562,
-0.06341552734375,
0.040435791015625,
0.021697998046875,
0.017547607421875,
0.01468658447265625,
-0.038604736328125,
-0.08892822265625,
0.1195068359375,
0.017974853515625,
-0.0550537109375,
-0.01442718505859375,
0.006103515625,
0.042694091796875,
-0.00836181640625,
0.03887939453125,
0.039947509765625,
0.034698486328125,
0.0012483596801757812,
-0.079345703125,
0.02044677734375,
-0.02239990234375,
0.008270263671875,
0.0220794677734375,
-0.0814208984375,
0.08953857421875,
-0.0106201171875,
-0.0019969940185546875,
0.0095367431640625,
0.0552978515625,
0.040771484375,
0.01194000244140625,
0.042083740234375,
0.0623779296875,
0.0628662109375,
-0.00678253173828125,
0.08697509765625,
-0.044464111328125,
0.054656982421875,
0.0670166015625,
0.0029964447021484375,
0.055084228515625,
0.031524658203125,
-0.0321044921875,
0.046966552734375,
0.07025146484375,
-0.01332855224609375,
0.0204620361328125,
0.01983642578125,
-0.0040435791015625,
-0.0073089599609375,
0.01499176025390625,
-0.045928955078125,
0.01090240478515625,
0.020721435546875,
-0.039093017578125,
-0.00884246826171875,
-0.0022563934326171875,
0.0201263427734375,
-0.01366424560546875,
-0.03094482421875,
0.0294342041015625,
0.012298583984375,
-0.045166015625,
0.06829833984375,
0.00817108154296875,
0.052764892578125,
-0.0386962890625,
0.0229034423828125,
-0.01198577880859375,
0.0158538818359375,
-0.026031494140625,
-0.04937744140625,
0.0075531005859375,
0.001636505126953125,
-0.0028629302978515625,
-0.01499176025390625,
0.040435791015625,
-0.0165863037109375,
-0.037506103515625,
0.031036376953125,
0.0269927978515625,
0.01470184326171875,
-0.012237548828125,
-0.07171630859375,
-0.00907135009765625,
0.006031036376953125,
-0.065673828125,
0.0313720703125,
0.025543212890625,
-0.006496429443359375,
0.045135498046875,
0.0455322265625,
-0.0023021697998046875,
0.0087127685546875,
0.0093536376953125,
0.07421875,
-0.04559326171875,
-0.0316162109375,
-0.06402587890625,
0.050506591796875,
-0.0015020370483398438,
-0.041900634765625,
0.055084228515625,
0.04949951171875,
0.05841064453125,
0.01152801513671875,
0.04736328125,
-0.023193359375,
0.0181427001953125,
-0.044219970703125,
0.05126953125,
-0.04437255859375,
0.01103973388671875,
-0.020263671875,
-0.07305908203125,
-0.00878143310546875,
0.043487548828125,
-0.03546142578125,
0.03448486328125,
0.059326171875,
0.06353759765625,
0.004718780517578125,
0.0041961669921875,
0.003772735595703125,
0.0226593017578125,
0.022857666015625,
0.0635986328125,
0.035430908203125,
-0.06488037109375,
0.057769775390625,
-0.0309906005859375,
-0.015411376953125,
-0.0108489990234375,
-0.050811767578125,
-0.05718994140625,
-0.039093017578125,
-0.032562255859375,
-0.032196044921875,
-0.0022735595703125,
0.05859375,
0.052764892578125,
-0.0504150390625,
-0.0185546875,
-0.030364990234375,
-0.01412200927734375,
-0.0166778564453125,
-0.0206756591796875,
0.05078125,
-0.0195770263671875,
-0.055999755859375,
0.005443572998046875,
-0.007343292236328125,
0.021820068359375,
-0.022369384765625,
-0.0279083251953125,
-0.0146636962890625,
-0.000022292137145996094,
0.02557373046875,
0.02520751953125,
-0.04339599609375,
-0.0170135498046875,
-0.003849029541015625,
-0.0243988037109375,
0.00846099853515625,
0.034149169921875,
-0.046844482421875,
-0.00003814697265625,
0.033721923828125,
0.02191162109375,
0.0718994140625,
-0.0091094970703125,
0.01552581787109375,
-0.03717041015625,
0.017303466796875,
0.008209228515625,
0.04217529296875,
0.016876220703125,
-0.03143310546875,
0.049652099609375,
0.0289459228515625,
-0.05859375,
-0.0609130859375,
-0.00626373291015625,
-0.07257080078125,
-0.0153045654296875,
0.0836181640625,
-0.0111083984375,
-0.02862548828125,
0.0178070068359375,
-0.01349639892578125,
0.028839111328125,
-0.0186767578125,
0.04541015625,
0.05218505859375,
-0.00372314453125,
-0.01346588134765625,
-0.05377197265625,
0.028045654296875,
0.04046630859375,
-0.053924560546875,
-0.0005092620849609375,
0.021453857421875,
0.031463623046875,
0.01427459716796875,
0.050262451171875,
-0.0234375,
0.0158843994140625,
0.00858306884765625,
0.0210418701171875,
-0.0013895034790039062,
-0.005764007568359375,
-0.0416259765625,
0.01093292236328125,
-0.00496673583984375,
-0.004253387451171875
]
] |
TheBloke/Genz-70b-GPTQ | 2023-09-27T12:46:24.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"en",
"license:llama2",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/Genz-70b-GPTQ | 35 | 7,463 | transformers | 2023-08-26T16:33:48 | ---
language:
- en
license: llama2
library_name: transformers
model_name: GenZ 70B
base_model: budecosystem/genz-70b
inference: false
model_creator: Bud
model_type: llama
pipeline_tag: text-generation
prompt_template: '### User:
{prompt}
### Assistant:
'
quantized_by: TheBloke
---
<!-- 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 -->
# GenZ 70B - GPTQ
- Model creator: [Bud](https://huggingface.co/budecosystem)
- Original model: [GenZ 70B](https://huggingface.co/budecosystem/genz-70b)
<!-- description start -->
## Description
This repo contains GPTQ model files for [Bud's GenZ 70B](https://huggingface.co/budecosystem/genz-70b).
Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
<!-- description end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Genz-70b-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Genz-70b-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Genz-70b-GGUF)
* [Bud's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/budecosystem/genz-70b)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: User-Assistant-Newlines
```
### User:
{prompt}
### Assistant:
```
<!-- prompt-template end -->
<!-- README_GPTQ.md-provided-files start -->
## Provided files and GPTQ parameters
Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.
Each separate quant is in a different branch. See below for instructions on fetching from different branches.
All recent GPTQ files are made with AutoGPTQ, and all files in non-main branches are made with AutoGPTQ. Files in the `main` branch which were uploaded before August 2023 were made with GPTQ-for-LLaMa.
<details>
<summary>Explanation of GPTQ parameters</summary>
- Bits: The bit size of the quantised model.
- GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
- Act Order: True or False. Also known as `desc_act`. True results in better quantisation accuracy. Some GPTQ clients have had issues with models that use Act Order plus Group Size, but this is generally resolved now.
- Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
- GPTQ dataset: The dataset used for quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
- Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
- ExLlama Compatibility: Whether this file can be loaded with ExLlama, which currently only supports Llama models in 4-bit.
</details>
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
| [main](https://huggingface.co/TheBloke/Genz-70b-GPTQ/tree/main) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 36.65 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/Genz-70b-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 40.66 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
| [gptq-3bit--1g-actorder_True](https://huggingface.co/TheBloke/Genz-70b-GPTQ/tree/gptq-3bit--1g-actorder_True) | 3 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 26.77 GB | No | 3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g. |
| [gptq-3bit-128g-actorder_True](https://huggingface.co/TheBloke/Genz-70b-GPTQ/tree/gptq-3bit-128g-actorder_True) | 3 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 28.03 GB | No | 3-bit, with group size 128g and act-order. Higher quality than 128g-False. |
<!-- README_GPTQ.md-provided-files end -->
<!-- README_GPTQ.md-download-from-branches start -->
## How to download from branches
- In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/Genz-70b-GPTQ:main`
- With Git, you can clone a branch with:
```
git clone --single-branch --branch main https://huggingface.co/TheBloke/Genz-70b-GPTQ
```
- In Python Transformers code, the branch is the `revision` parameter; see below.
<!-- README_GPTQ.md-download-from-branches end -->
<!-- README_GPTQ.md-text-generation-webui start -->
## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.
1. Click the **Model tab**.
2. Under **Download custom model or LoRA**, enter `TheBloke/Genz-70b-GPTQ`.
- To download from a specific branch, enter for example `TheBloke/Genz-70b-GPTQ:main`
- see Provided Files above for the list of branches for each option.
3. Click **Download**.
4. The model will start downloading. Once it's finished it will say "Done".
5. In the top left, click the refresh icon next to **Model**.
6. In the **Model** dropdown, choose the model you just downloaded: `Genz-70b-GPTQ`
7. The model will automatically load, and is now ready for use!
8. If you want any custom settings, set them and then click **Save settings for this model** followed by **Reload the Model** in the top right.
* Note that you do not need to and should not set manual GPTQ parameters any more. These are set automatically from the file `quantize_config.json`.
9. Once you're ready, click the **Text Generation tab** and enter a prompt to get started!
<!-- README_GPTQ.md-text-generation-webui end -->
<!-- README_GPTQ.md-use-from-python start -->
## How to use this GPTQ model from Python code
### Install the necessary packages
Requires: Transformers 4.32.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.
```shell
pip3 install transformers>=4.32.0 optimum>=1.12.0
pip3 install auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/ # Use cu117 if on CUDA 11.7
```
If you have problems installing AutoGPTQ using the pre-built wheels, install it from source instead:
```shell
pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
pip3 install .
```
### For CodeLlama models only: you must use Transformers 4.33.0 or later.
If 4.33.0 is not yet released when you read this, you will need to install Transformers from source:
```shell
pip3 uninstall -y transformers
pip3 install git+https://github.com/huggingface/transformers.git
```
### You can then use the following code
```python
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
model_name_or_path = "TheBloke/Genz-70b-GPTQ"
# To use a different branch, change revision
# For example: revision="main"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
device_map="auto",
trust_remote_code=False,
revision="main")
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
prompt = "Tell me about AI"
prompt_template=f'''### User:
{prompt}
### Assistant:
'''
print("\n\n*** Generate:")
input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
print(tokenizer.decode(output[0]))
# Inference can also be done using transformers' pipeline
print("*** Pipeline:")
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
max_new_tokens=512,
do_sample=True,
temperature=0.7,
top_p=0.95,
top_k=40,
repetition_penalty=1.1
)
print(pipe(prompt_template)[0]['generated_text'])
```
<!-- README_GPTQ.md-use-from-python end -->
<!-- README_GPTQ.md-compatibility start -->
## Compatibility
The files provided are tested to work with AutoGPTQ, both via Transformers and using AutoGPTQ directly. They should also work with [Occ4m's GPTQ-for-LLaMa fork](https://github.com/0cc4m/KoboldAI).
[ExLlama](https://github.com/turboderp/exllama) is compatible with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
[Huggingface Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) is compatible with all GPTQ models.
<!-- README_GPTQ.md-compatibility end -->
<!-- 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!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
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**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
# Original model card: Bud's GenZ 70B
---
<div align="center"><h1 align="center">~ GenZ ~</h1><img src="https://raw.githubusercontent.com/BudEcosystem/GenZ/main/assets/genz-logo.png" width=150></div>
<p align="center"><i>Democratizing access to LLMs for the open-source community.<br>Let's advance AI, together. </i></p>
---
## Introduction 🎉
Welcome to **GenZ**, an advanced Large Language Model (LLM) fine-tuned on the foundation of Meta's open-source Llama V2 70B parameter model. At Bud Ecosystem, we believe in the power of open-source collaboration to drive the advancement of technology at an accelerated pace. Our vision is to democratize access to fine-tuned LLMs, and to that end, we will be releasing a series of models across different parameter counts (7B, 13B, and 70B) and quantizations (32-bit and 4-bit) for the open-source community to use, enhance, and build upon.
<p align="center"><img src="https://raw.githubusercontent.com/BudEcosystem/GenZ/main/assets/mt_bench_compare.png" width="500"></p>
The smaller quantization version of our models makes them more accessible, enabling their use even on personal computers. This opens up a world of possibilities for developers, researchers, and enthusiasts to experiment with these models and contribute to the collective advancement of language model technology.
GenZ isn't just a powerful text generator—it's a sophisticated AI assistant, capable of understanding and responding to user prompts with high-quality responses. We've taken the robust capabilities of Llama V2 and fine-tuned them to offer a more user-focused experience. Whether you're seeking informative responses or engaging interactions, GenZ is designed to deliver.
And this isn't the end. It's just the beginning of a journey towards creating more advanced, more efficient, and more accessible language models. We invite you to join us on this exciting journey. 🚀
---
<h2>Milestone Releases ️🏁</h2>
**[21 August 2023]**
[_GenZ-70B_](https://huggingface.co/budecosystem/genz-70b) : We're excited to announce the release of our Genz 70BB model. Experience the advancements by downloading the model from [HuggingFace](https://huggingface.co/budecosystem/genz-70b).
**[27 July 2023]**
[_GenZ-13B V2 (ggml)_](https://huggingface.co/budecosystem/genz-13b-v2-ggml) : Announcing our GenZ-13B v2 with ggml. This variant of GenZ can run inferencing using only CPU and without the need of GPU. Download the model from [HuggingFace](https://huggingface.co/budecosystem/genz-13b-v2-ggml).
**[27 July 2023]**
[_GenZ-13B V2 (4-bit)_](https://huggingface.co/budecosystem/genz-13b-v2-4bit) : Announcing our GenZ-13B v2 with 4-bit quantisation. Enabling inferencing with much lesser GPU memory than the 32-bit variant. Download the model from [HuggingFace](https://huggingface.co/budecosystem/genz-13b-v2-4bit).
**[26 July 2023]**
[_GenZ-13B V2_](https://huggingface.co/budecosystem/genz-13b-v2) : We're excited to announce the release of our Genz 13B v2 model, a step forward with improved evaluation results compared to v1. Experience the advancements by downloading the model from [HuggingFace](https://huggingface.co/budecosystem/genz-13b-v2).
**[20 July 2023]**
[_GenZ-13B_](https://huggingface.co/budecosystem/genz-13b) : We marked an important milestone with the release of the Genz 13B model. The journey began here, and you can partake in it by downloading the model from [Hugging Face](https://huggingface.co/budecosystem/genz-13b).
---
<h2>Evaluations 🎯</h2>
Evaluating our model is a key part of our fine-tuning process. It helps us understand how our model is performing and how it stacks up against other models. Here's a look at some of the key evaluations for GenZ 70B:
<h3>Benchmark Comparison</h3>
We've compared GenZ models to understand the improvements our fine-tuning has achieved.
| Model Name | MT Bench | MMLU | Human Eval | BBH |
|:----------:|:--------:|:----:|:----------:|:----:|
| Genz 13B | 6.12 | 53.62| 17.68 | 37.76|
| Genz 13B v2| 6.79 | 53.68| 21.95 | 38.1 |
| Genz 70B | 7.33 | 70.32| 37.8 |54.69 |
<h3>MT Bench Score</h3>
A key evaluation metric we use is the MT Bench score. This score provides a comprehensive assessment of our model's performance across a range of tasks.
<p align="center"><img src="https://raw.githubusercontent.com/BudEcosystem/GenZ/main/assets/mt_bench_score.png" width="500"></p>
---
<h2>Getting Started on Hugging Face 🤗</h2>
Getting up and running with our models on Hugging Face is a breeze. Follow these steps:
<h3>1️⃣ : Import necessary modules</h3>
Start by importing the necessary modules from the ‘transformers’ library and ‘torch’.
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("budecosystem/genz-70b", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("budecosystem/genz-70b", torch_dtype=torch.bfloat16, rope_scaling={"type": "dynamic", "factor": 2})
prompt = "### User:\nWrite a python flask code for login management\n\n### Assistant:\n"
inputs = tokenizer(prompt, return_tensors="pt")
sample = model.generate(**inputs, max_length=128)
print(tokenizer.decode(sample[0]))
```
Want to interact with the model in a more intuitive way? We have a Gradio interface set up for that. Head over to our GitHub page, clone the repository, and run the ‘generate.py’ script to try it out. Happy experimenting! 😄
<h2>Why Use GenZ? 💡</h2>
You might be wondering, "Why should I choose GenZ over a pretrained model?" The answer lies in the extra mile we've gone to fine-tune our models.
While pretrained models are undeniably powerful, GenZ brings something extra to the table. We've fine-tuned it with curated datasets, which means it has additional skills and capabilities beyond what a pretrained model can offer. Whether you need it for a simple task or a complex project, GenZ is up for the challenge.
What's more, we are committed to continuously enhancing GenZ. We believe in the power of constant learning and improvement. That's why we'll be regularly fine-tuning our models with various curated datasets to make them even better. Our goal is to reach the state of the art and beyond - and we're committed to staying the course until we get there.
But don't just take our word for it. We've provided detailed evaluations and performance details in a later section, so you can see the difference for yourself.
Choose GenZ and join us on this journey. Together, we can push the boundaries of what's possible with large language models.
---
<h2>Model Card for GenZ 70B 📄</h2>
Here's a quick overview of everything you need to know about GenZ 70B.
<h3>Model Details:</h3>
- Developed by: Bud Ecosystem
- Base pretrained model type: Llama V2 70B
- Model Architecture: GenZ 70B, fine-tuned on Llama V2 70B, is an auto-regressive language model that employs an optimized transformer architecture. The fine-tuning process for GenZ 70B leveraged Supervised Fine-Tuning (SFT)
- License: The model is available for commercial use under a custom commercial license. For more information, please visit: [Meta AI Model and Library Downloads](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
---
<h2>Intended Use 💼</h2>
When we created GenZ 70B, we had a clear vision of how it could be used to push the boundaries of what's possible with large language models. We also understand the importance of using such models responsibly. Here's a brief overview of the intended and out-of-scope uses for GenZ 70B.
<h3>Direct Use</h3>
GenZ 70B is designed to be a powerful tool for research on large language models. It's also an excellent foundation for further specialization and fine-tuning for specific use cases, such as:
- Text summarization
- Text generation
- Chatbot creation
- And much more!
<h3>Out-of-Scope Use 🚩</h3>
While GenZ 70B is versatile, there are certain uses that are out of scope:
- Production use without adequate assessment of risks and mitigation
- Any use cases which may be considered irresponsible or harmful
- Use in any manner that violates applicable laws or regulations, including trade compliance laws
- Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Llama 2
Remember, GenZ 70B, like any large language model, is trained on a large-scale corpora representative of the web, and therefore, may carry the stereotypes and biases commonly encountered online.
<h3>Recommendations 🧠</h3>
We recommend users of GenZ 70B to consider fine-tuning it for the specific set of tasks of interest. Appropriate precautions and guardrails should be taken for any production use. Using GenZ 70B responsibly is key to unlocking its full potential while maintaining a safe and respectful environment.
---
<h2>Training Details 📚</h2>
When fine-tuning GenZ 70B, we took a meticulous approach to ensure we were building on the solid base of the pretrained Llama V2 70B model in the most effective way. Here's a look at the key details of our training process:
<h3>Fine-Tuning Training Data</h3>
For the fine-tuning process, we used a carefully curated mix of datasets. These included data from OpenAssistant, an instruction fine-tuning dataset, and Thought Source for the Chain Of Thought (CoT) approach. This diverse mix of data sources helped us enhance the model's capabilities across a range of tasks.
<h3>Hyperparameters</h3>
Here are the hyperparameters we used for fine-tuning:
| Hyperparameter | Value |
| -------------- | ----- |
| Warmup Ratio | 0.04 |
| Learning Rate Scheduler Type | Cosine |
| Learning Rate | 2e-5 |
| Number of Training Epochs | 3 |
| Per Device Training Batch Size | 4 |
| Gradient Accumulation Steps | 4 |
| Precision | FP16 |
| Optimizer | AdamW |
---
<h2>Looking Ahead 👀</h2>
We're excited about the journey ahead with GenZ. We're committed to continuously improving and enhancing our models, and we're excited to see what the open-source community will build with them. We believe in the power of collaboration, and we can't wait to see what we can achieve together.
Remember, we're just getting started. This is just the beginning of a journey that we believe will revolutionize the world of large language models. We invite you to join us on this exciting journey. Together, we can push the boundaries of what's possible with AI. 🚀
---
Check the GitHub for the code -> [GenZ](https://raw.githubusercontent.com/BudEcosystem/GenZ)
| 24,172 | [
[
-0.03955078125,
-0.06109619140625,
0.01300811767578125,
0.01432037353515625,
-0.0230712890625,
-0.0078277587890625,
0.00867462158203125,
-0.034515380859375,
0.0103607177734375,
0.02777099609375,
-0.0487060546875,
-0.038055419921875,
-0.0229949951171875,
0.0010585784912109375,
-0.0291748046875,
0.0772705078125,
0.0057525634765625,
-0.0222625732421875,
-0.0035228729248046875,
-0.00679779052734375,
-0.0190887451171875,
-0.025421142578125,
-0.056671142578125,
-0.0215606689453125,
0.0230712890625,
0.0087890625,
0.06158447265625,
0.03985595703125,
0.01488494873046875,
0.025634765625,
-0.006519317626953125,
-0.004428863525390625,
-0.036529541015625,
-0.007251739501953125,
0.01044464111328125,
-0.016815185546875,
-0.049407958984375,
0.009185791015625,
0.0311431884765625,
0.0128173828125,
-0.0308380126953125,
0.015289306640625,
0.001888275146484375,
0.052764892578125,
-0.03387451171875,
0.01239013671875,
-0.0307464599609375,
0.00506591796875,
-0.00598907470703125,
0.007610321044921875,
-0.011016845703125,
-0.03912353515625,
0.0095977783203125,
-0.06805419921875,
0.024139404296875,
-0.004344940185546875,
0.094482421875,
0.00690460205078125,
-0.05035400390625,
0.01425933837890625,
-0.039947509765625,
0.0467529296875,
-0.07354736328125,
0.024871826171875,
0.035247802734375,
0.0203399658203125,
-0.01715087890625,
-0.072021484375,
-0.04327392578125,
-0.005565643310546875,
-0.0077362060546875,
0.025604248046875,
-0.037384033203125,
0.006855010986328125,
0.03485107421875,
0.056121826171875,
-0.0684814453125,
-0.01873779296875,
-0.0282745361328125,
-0.01229095458984375,
0.0634765625,
0.015045166015625,
0.0283050537109375,
-0.02117919921875,
-0.027557373046875,
-0.03424072265625,
-0.053802490234375,
0.005916595458984375,
0.0283050537109375,
0.0035800933837890625,
-0.0460205078125,
0.033294677734375,
-0.0294036865234375,
0.03924560546875,
0.01678466796875,
-0.00580596923828125,
0.027374267578125,
-0.042388916015625,
-0.03570556640625,
-0.028076171875,
0.09857177734375,
0.0272064208984375,
-0.0123443603515625,
0.0189971923828125,
-0.005039215087890625,
-0.01554107666015625,
0.008331298828125,
-0.0780029296875,
-0.03759765625,
0.03515625,
-0.03692626953125,
-0.0234832763671875,
-0.004566192626953125,
-0.05975341796875,
0.0019989013671875,
-0.00543212890625,
0.0423583984375,
-0.04791259765625,
-0.0273895263671875,
0.0075225830078125,
-0.035247802734375,
0.03363037109375,
0.02337646484375,
-0.0592041015625,
0.0290679931640625,
0.0187225341796875,
0.05169677734375,
0.0172576904296875,
-0.00771331787109375,
-0.0157012939453125,
0.010650634765625,
-0.01004791259765625,
0.03497314453125,
-0.0091705322265625,
-0.040679931640625,
-0.019256591796875,
0.02117919921875,
0.0084228515625,
-0.020263671875,
0.037994384765625,
-0.0236968994140625,
0.034149169921875,
-0.03265380859375,
-0.03887939453125,
-0.030975341796875,
0.0055999755859375,
-0.049774169921875,
0.09356689453125,
0.041259765625,
-0.06475830078125,
0.016510009765625,
-0.0462646484375,
-0.020416259765625,
0.00843048095703125,
0.0014066696166992188,
-0.047882080078125,
-0.005809783935546875,
0.015960693359375,
0.0220947265625,
-0.022735595703125,
0.00772857666015625,
-0.021759033203125,
-0.020050048828125,
0.01149749755859375,
-0.04132080078125,
0.10015869140625,
0.0182647705078125,
-0.03692626953125,
-0.003833770751953125,
-0.06005859375,
0.012298583984375,
0.037200927734375,
-0.0228118896484375,
-0.0004911422729492188,
-0.011566162109375,
0.00811004638671875,
0.00489044189453125,
0.0172882080078125,
-0.0279083251953125,
0.0350341796875,
-0.01885986328125,
0.05560302734375,
0.04608154296875,
0.004589080810546875,
0.01715087890625,
-0.03131103515625,
0.041229248046875,
-0.0016031265258789062,
0.046600341796875,
0.006439208984375,
-0.054412841796875,
-0.052398681640625,
-0.01340484619140625,
0.033843994140625,
0.0450439453125,
-0.0584716796875,
0.038970947265625,
-0.01218414306640625,
-0.05572509765625,
-0.0186004638671875,
-0.0092010498046875,
0.020294189453125,
0.0229644775390625,
0.03363037109375,
-0.029693603515625,
-0.028472900390625,
-0.0638427734375,
0.01097869873046875,
-0.037322998046875,
-0.00823211669921875,
0.0264739990234375,
0.05462646484375,
-0.018646240234375,
0.06451416015625,
-0.056732177734375,
-0.0042572021484375,
-0.00510406494140625,
0.01351165771484375,
0.0199432373046875,
0.044219970703125,
0.057342529296875,
-0.060516357421875,
-0.040191650390625,
-0.00846099853515625,
-0.048370361328125,
-0.00469970703125,
0.0030384063720703125,
-0.0287933349609375,
0.007671356201171875,
0.003849029541015625,
-0.08062744140625,
0.048126220703125,
0.03765869140625,
-0.0445556640625,
0.0654296875,
-0.020294189453125,
0.01251983642578125,
-0.08782958984375,
0.00559234619140625,
0.0161285400390625,
-0.0269775390625,
-0.038055419921875,
0.01506805419921875,
0.003620147705078125,
0.008575439453125,
-0.0330810546875,
0.04345703125,
-0.035430908203125,
0.00911712646484375,
0.006359100341796875,
-0.00799560546875,
0.028839111328125,
0.03564453125,
-0.0194549560546875,
0.0655517578125,
0.0287933349609375,
-0.03692626953125,
0.0443115234375,
0.033447265625,
-0.0076446533203125,
0.01751708984375,
-0.065185546875,
0.0099029541015625,
0.01123809814453125,
0.027618408203125,
-0.068603515625,
-0.0214691162109375,
0.041351318359375,
-0.0435791015625,
0.03515625,
-0.0260009765625,
-0.027618408203125,
-0.034698486328125,
-0.042633056640625,
0.0258331298828125,
0.0635986328125,
-0.025848388671875,
0.040924072265625,
0.025634765625,
-0.0013704299926757812,
-0.039459228515625,
-0.051055908203125,
-0.01409149169921875,
-0.0211334228515625,
-0.043426513671875,
0.041778564453125,
-0.013031005859375,
-0.004322052001953125,
0.0013494491577148438,
0.00913238525390625,
-0.008575439453125,
-0.0032176971435546875,
0.02191162109375,
0.026824951171875,
-0.01209259033203125,
-0.00994873046875,
0.017852783203125,
0.0035305023193359375,
-0.0003495216369628906,
-0.0231475830078125,
0.03662109375,
-0.01117706298828125,
0.001361846923828125,
-0.029571533203125,
0.0196685791015625,
0.0360107421875,
-0.0025634765625,
0.053619384765625,
0.06158447265625,
-0.027191162109375,
0.007843017578125,
-0.0355224609375,
-0.0085906982421875,
-0.03887939453125,
0.01389312744140625,
-0.0162200927734375,
-0.046600341796875,
0.035430908203125,
0.026702880859375,
0.01425933837890625,
0.061431884765625,
0.034576416015625,
-0.003307342529296875,
0.0777587890625,
0.035552978515625,
-0.00763702392578125,
0.038543701171875,
-0.04779052734375,
-0.01337432861328125,
-0.06121826171875,
-0.013153076171875,
-0.03192138671875,
-0.01116180419921875,
-0.059844970703125,
-0.0307769775390625,
0.0258026123046875,
0.021759033203125,
-0.062347412109375,
0.044097900390625,
-0.05267333984375,
0.0112457275390625,
0.047821044921875,
0.021636962890625,
0.019775390625,
0.00679779052734375,
-0.01096343994140625,
0.00992584228515625,
-0.043701171875,
-0.0223236083984375,
0.0831298828125,
0.024627685546875,
0.047576904296875,
0.015228271484375,
0.038238525390625,
0.01036834716796875,
0.020782470703125,
-0.04046630859375,
0.040496826171875,
-0.0020809173583984375,
-0.055816650390625,
-0.0272216796875,
-0.052398681640625,
-0.07086181640625,
0.015960693359375,
-0.0094451904296875,
-0.056671142578125,
0.027557373046875,
0.003726959228515625,
-0.0330810546875,
0.0192413330078125,
-0.05340576171875,
0.07305908203125,
-0.007572174072265625,
-0.0283355712890625,
0.00272369384765625,
-0.048248291015625,
0.0278472900390625,
0.01551055908203125,
0.0017652511596679688,
-0.01458740234375,
-0.00814056396484375,
0.057281494140625,
-0.06396484375,
0.05548095703125,
-0.01806640625,
-0.006755828857421875,
0.04241943359375,
-0.0100860595703125,
0.0428466796875,
0.00921630859375,
0.0037250518798828125,
0.023895263671875,
0.0228424072265625,
-0.03497314453125,
-0.0270538330078125,
0.036712646484375,
-0.07440185546875,
-0.03973388671875,
-0.0321044921875,
-0.03253173828125,
0.0011110305786132812,
0.00812530517578125,
0.045440673828125,
0.03363037109375,
-0.00861358642578125,
0.0016794204711914062,
0.050201416015625,
-0.0307159423828125,
0.034759521484375,
0.0229644775390625,
-0.0280609130859375,
-0.048858642578125,
0.06402587890625,
0.005504608154296875,
0.0194854736328125,
0.0186920166015625,
0.01326751708984375,
-0.03497314453125,
-0.037139892578125,
-0.059295654296875,
0.020050048828125,
-0.039520263671875,
-0.033233642578125,
-0.04486083984375,
-0.026397705078125,
-0.036346435546875,
0.0196685791015625,
-0.0268707275390625,
-0.047088623046875,
-0.0318603515625,
-0.0010128021240234375,
0.0684814453125,
0.039337158203125,
-0.007904052734375,
0.0256805419921875,
-0.06597900390625,
0.021453857421875,
0.03765869140625,
0.0153961181640625,
-0.00183868408203125,
-0.0521240234375,
-0.006076812744140625,
0.0211181640625,
-0.05621337890625,
-0.07476806640625,
0.04852294921875,
0.01334381103515625,
0.0269317626953125,
0.033660888671875,
0.012176513671875,
0.056884765625,
-0.0202178955078125,
0.07861328125,
0.019317626953125,
-0.06884765625,
0.0435791015625,
-0.045806884765625,
0.02105712890625,
0.032318115234375,
0.04351806640625,
-0.0206298828125,
-0.022613525390625,
-0.055755615234375,
-0.062042236328125,
0.031341552734375,
0.03753662109375,
0.007015228271484375,
0.01018524169921875,
0.04205322265625,
0.00044989585876464844,
0.011993408203125,
-0.07061767578125,
-0.043121337890625,
-0.0304107666015625,
-0.01470184326171875,
0.01422119140625,
-0.0033969879150390625,
-0.0191192626953125,
-0.050079345703125,
0.07623291015625,
-0.01148223876953125,
0.052276611328125,
0.0211181640625,
0.01456451416015625,
-0.01116180419921875,
0.014373779296875,
0.0229949951171875,
0.045989990234375,
-0.0185699462890625,
-0.0206756591796875,
0.01090240478515625,
-0.060516357421875,
0.01641845703125,
0.03204345703125,
-0.028106689453125,
-0.00606536865234375,
0.0025577545166015625,
0.06103515625,
-0.00659942626953125,
-0.02288818359375,
0.037872314453125,
-0.025634765625,
-0.03045654296875,
-0.02276611328125,
0.0165252685546875,
0.0186004638671875,
0.0294036865234375,
0.032257080078125,
-0.019378662109375,
0.0252685546875,
-0.042236328125,
0.0027923583984375,
0.037994384765625,
-0.01654052734375,
-0.0233306884765625,
0.06817626953125,
0.0013580322265625,
0.003520965576171875,
0.0609130859375,
-0.0291290283203125,
-0.034698486328125,
0.059844970703125,
0.036163330078125,
0.0623779296875,
-0.011322021484375,
0.0272979736328125,
0.044525146484375,
0.01177978515625,
-0.007965087890625,
0.0301361083984375,
0.0017900466918945312,
-0.04351806640625,
-0.026885986328125,
-0.04998779296875,
-0.0204010009765625,
0.02423095703125,
-0.055908203125,
0.0169219970703125,
-0.0307464599609375,
-0.036224365234375,
-0.0113067626953125,
0.0207366943359375,
-0.040985107421875,
0.02520751953125,
0.0009417533874511719,
0.0699462890625,
-0.05291748046875,
0.06353759765625,
0.040771484375,
-0.039306640625,
-0.075439453125,
-0.01078033447265625,
0.00293731689453125,
-0.041900634765625,
0.012115478515625,
0.00290679931640625,
0.0255584716796875,
0.00994873046875,
-0.05615234375,
-0.06329345703125,
0.11004638671875,
0.0262298583984375,
-0.04632568359375,
-0.01380157470703125,
-0.0002205371856689453,
0.0241851806640625,
-0.0016155242919921875,
0.04925537109375,
0.038543701171875,
0.0278778076171875,
0.00984954833984375,
-0.0638427734375,
0.0357666015625,
-0.03271484375,
0.0041656494140625,
0.022491455078125,
-0.07568359375,
0.0765380859375,
0.0003974437713623047,
-0.01383209228515625,
0.01849365234375,
0.045654296875,
0.028961181640625,
0.0002009868621826172,
0.03009033203125,
0.060699462890625,
0.051666259765625,
-0.030914306640625,
0.08489990234375,
-0.0168914794921875,
0.050994873046875,
0.05859375,
0.00553131103515625,
0.055084228515625,
0.01531219482421875,
-0.061248779296875,
0.048736572265625,
0.07269287109375,
-0.01456451416015625,
0.030914306640625,
0.005126953125,
-0.026336669921875,
-0.009033203125,
0.0144195556640625,
-0.0584716796875,
0.004764556884765625,
0.02435302734375,
-0.019622802734375,
0.0069427490234375,
-0.01456451416015625,
-0.00606536865234375,
-0.050689697265625,
-0.01308441162109375,
0.042388916015625,
0.019195556640625,
-0.019989013671875,
0.061248779296875,
-0.0037631988525390625,
0.046905517578125,
-0.04608154296875,
-0.01407623291015625,
-0.0269775390625,
-0.0093994140625,
-0.019683837890625,
-0.0521240234375,
0.006359100341796875,
-0.0144500732421875,
-0.01190185546875,
0.0019006729125976562,
0.052947998046875,
-0.0137786865234375,
-0.038726806640625,
0.0213470458984375,
0.034515380859375,
0.01751708984375,
-0.01047515869140625,
-0.078369140625,
0.01047515869140625,
0.0033283233642578125,
-0.05169677734375,
0.030975341796875,
0.0291290283203125,
0.0181884765625,
0.05145263671875,
0.044403076171875,
-0.008392333984375,
0.006641387939453125,
-0.0069427490234375,
0.07098388671875,
-0.062042236328125,
-0.022918701171875,
-0.06158447265625,
0.049530029296875,
-0.01000213623046875,
-0.0325927734375,
0.053253173828125,
0.0460205078125,
0.056182861328125,
0.0006365776062011719,
0.059844970703125,
-0.03057861328125,
0.0101318359375,
-0.0285186767578125,
0.058929443359375,
-0.05645751953125,
0.00664520263671875,
-0.02874755859375,
-0.0592041015625,
0.005863189697265625,
0.05279541015625,
-0.0027446746826171875,
0.02349853515625,
0.029296875,
0.059051513671875,
0.0012578964233398438,
0.0157928466796875,
0.01364898681640625,
0.0291748046875,
0.01172637939453125,
0.061431884765625,
0.057159423828125,
-0.076904296875,
0.0457763671875,
-0.0340576171875,
-0.01453399658203125,
-0.0010538101196289062,
-0.0592041015625,
-0.051605224609375,
-0.034423828125,
-0.046112060546875,
-0.0457763671875,
-0.0028171539306640625,
0.06707763671875,
0.06170654296875,
-0.047882080078125,
-0.0189971923828125,
-0.01514434814453125,
-0.0032482147216796875,
-0.0194244384765625,
-0.024688720703125,
0.0224761962890625,
0.0190277099609375,
-0.05767822265625,
0.015045166015625,
0.0016021728515625,
0.036956787109375,
-0.0135955810546875,
-0.016448974609375,
-0.0181884765625,
-0.0020771026611328125,
0.042724609375,
0.033233642578125,
-0.0386962890625,
-0.006633758544921875,
-0.01438140869140625,
-0.005558013916015625,
0.0222015380859375,
0.02532958984375,
-0.058807373046875,
0.0011377334594726562,
0.03887939453125,
0.00864410400390625,
0.0631103515625,
0.0003654956817626953,
0.03887939453125,
-0.035400390625,
0.0021381378173828125,
0.002349853515625,
0.0290679931640625,
0.007427215576171875,
-0.0396728515625,
0.049285888671875,
0.0297088623046875,
-0.049407958984375,
-0.051788330078125,
-0.0100250244140625,
-0.0826416015625,
-0.023406982421875,
0.08294677734375,
-0.0110931396484375,
-0.0243682861328125,
-0.0026035308837890625,
-0.0190887451171875,
0.037078857421875,
-0.03985595703125,
0.029693603515625,
0.0311737060546875,
-0.0126190185546875,
-0.0222625732421875,
-0.0604248046875,
0.041107177734375,
0.0129852294921875,
-0.06597900390625,
-0.002105712890625,
0.03619384765625,
0.040008544921875,
0.0020122528076171875,
0.0592041015625,
-0.015625,
0.027679443359375,
0.0124969482421875,
0.0096588134765625,
0.0024280548095703125,
0.01042938232421875,
-0.023895263671875,
0.0038776397705078125,
-0.0178070068359375,
-0.0011720657348632812
]
] |
SCUT-DLVCLab/lilt-roberta-en-base | 2023-08-31T07:59:36.000Z | [
"transformers",
"pytorch",
"safetensors",
"lilt",
"feature-extraction",
"vision",
"arxiv:2202.13669",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | feature-extraction | SCUT-DLVCLab | null | null | SCUT-DLVCLab/lilt-roberta-en-base | 14 | 7,458 | transformers | 2022-09-29T14:06:32 | ---
license: mit
tags:
- vision
---
# LiLT-RoBERTa (base-sized model)
Language-Independent Layout Transformer - RoBERTa model by stitching a pre-trained RoBERTa (English) and a pre-trained Language-Independent Layout Transformer (LiLT) together. It was introduced in the paper [LiLT: A Simple yet Effective Language-Independent Layout Transformer for Structured Document Understanding](https://arxiv.org/abs/2202.13669) by Wang et al. and first released in [this repository](https://github.com/jpwang/lilt).
Disclaimer: The team releasing LiLT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Language-Independent Layout Transformer (LiLT) allows to combine any pre-trained RoBERTa encoder from the hub (hence, in any language) with a lightweight Layout Transformer to have a LayoutLM-like model for any language.
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/lilt_architecture.jpg" alt="drawing" width="600"/>
## Intended uses & limitations
The model is meant to be fine-tuned on tasks like document image classification, document parsing and document QA. See the [model hub](https://huggingface.co/models?search=lilt) to look for fine-tuned versions on a task that interests you.
### How to use
For code examples, we refer to the [documentation](https://huggingface.co/transformers/main/model_doc/lilt.html).
### BibTeX entry and citation info
```bibtex
@misc{https://doi.org/10.48550/arxiv.2202.13669,
doi = {10.48550/ARXIV.2202.13669},
url = {https://arxiv.org/abs/2202.13669},
author = {Wang, Jiapeng and Jin, Lianwen and Ding, Kai},
keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {LiLT: A Simple yet Effective Language-Independent Layout Transformer for Structured Document Understanding},
publisher = {arXiv},
year = {2022},
copyright = {arXiv.org perpetual, non-exclusive license}
}
``` | 2,078 | [
[
-0.029266357421875,
-0.06036376953125,
0.0240325927734375,
0.0274658203125,
-0.005466461181640625,
-0.0183258056640625,
-0.007732391357421875,
-0.0293731689453125,
0.0284271240234375,
0.0183563232421875,
-0.050994873046875,
-0.024261474609375,
-0.050994873046875,
-0.00643157958984375,
-0.03594970703125,
0.10638427734375,
-0.00782012939453125,
-0.0027828216552734375,
-0.019622802734375,
-0.0195159912109375,
-0.01313018798828125,
-0.017242431640625,
-0.03167724609375,
-0.0275421142578125,
0.024505615234375,
0.00946807861328125,
0.05926513671875,
0.051910400390625,
0.052276611328125,
0.030853271484375,
-0.0202484130859375,
-0.00635528564453125,
-0.01459503173828125,
-0.014495849609375,
0.004985809326171875,
-0.03997802734375,
-0.06280517578125,
0.00014293193817138672,
0.05145263671875,
0.034881591796875,
0.007434844970703125,
0.021484375,
0.00678253173828125,
0.045867919921875,
-0.0113677978515625,
0.0266571044921875,
-0.017181396484375,
0.002239227294921875,
-0.017669677734375,
0.00437164306640625,
-0.0298614501953125,
-0.031494140625,
0.004638671875,
-0.03472900390625,
0.0270233154296875,
-0.00286865234375,
0.100830078125,
0.0172271728515625,
-0.01517486572265625,
-0.0195159912109375,
-0.041229248046875,
0.049407958984375,
-0.03228759765625,
0.04913330078125,
0.0009870529174804688,
0.01617431640625,
0.004283905029296875,
-0.07421875,
-0.052886962890625,
-0.017547607421875,
-0.038848876953125,
0.021453857421875,
-0.0204925537109375,
0.0005970001220703125,
0.04473876953125,
0.0477294921875,
-0.0638427734375,
-0.019805908203125,
-0.03533935546875,
-0.01296234130859375,
0.026702880859375,
-0.01282501220703125,
0.06561279296875,
-0.041259765625,
-0.04254150390625,
-0.0161285400390625,
-0.02825927734375,
0.00461578369140625,
0.0128021240234375,
0.01464080810546875,
-0.05694580078125,
0.033294677734375,
0.0159912109375,
0.044036865234375,
0.0201263427734375,
-0.03173828125,
0.027862548828125,
-0.010986328125,
-0.0225677490234375,
-0.035614013671875,
0.079345703125,
-0.004604339599609375,
-0.0009150505065917969,
-0.00994873046875,
-0.018524169921875,
-0.001056671142578125,
0.01605224609375,
-0.0670166015625,
-0.0338134765625,
0.007076263427734375,
-0.038665771484375,
-0.0122528076171875,
0.0033283233642578125,
-0.038116455078125,
-0.0019426345825195312,
-0.01922607421875,
0.029144287109375,
-0.045318603515625,
-0.018157958984375,
-0.0233001708984375,
-0.0012083053588867188,
0.028411865234375,
0.0282135009765625,
-0.0699462890625,
0.0215911865234375,
0.04022216796875,
0.06951904296875,
-0.0018968582153320312,
-0.02838134765625,
-0.0302734375,
-0.00582122802734375,
-0.00159454345703125,
0.06854248046875,
-0.0036449432373046875,
-0.03131103515625,
-0.01177978515625,
0.01184844970703125,
-0.01029205322265625,
-0.0197906494140625,
0.0706787109375,
-0.0341796875,
0.039764404296875,
-0.0002256631851196289,
-0.0254974365234375,
-0.0029296875,
0.0276031494140625,
-0.0567626953125,
0.06610107421875,
0.036590576171875,
-0.07855224609375,
0.00829315185546875,
-0.056182861328125,
-0.0211181640625,
0.007244110107421875,
-0.0222015380859375,
-0.051788330078125,
-0.007415771484375,
0.01015472412109375,
0.011199951171875,
-0.0092010498046875,
-0.004360198974609375,
-0.0010652542114257812,
-0.01384735107421875,
0.0028438568115234375,
-0.0089263916015625,
0.0872802734375,
0.01325225830078125,
-0.0032367706298828125,
0.03167724609375,
-0.054046630859375,
0.006107330322265625,
0.024261474609375,
-0.0307464599609375,
-0.0162200927734375,
-0.031829833984375,
0.03369140625,
0.015869140625,
0.03546142578125,
-0.03790283203125,
0.042205810546875,
-0.035858154296875,
0.021026611328125,
0.03533935546875,
-0.028656005859375,
0.0611572265625,
-0.034332275390625,
0.0537109375,
-0.0250396728515625,
0.0252532958984375,
-0.0263824462890625,
-0.04266357421875,
-0.06280517578125,
-0.03607177734375,
0.03594970703125,
0.049835205078125,
-0.044830322265625,
0.026458740234375,
-0.0267791748046875,
-0.040771484375,
-0.05584716796875,
0.0169677734375,
0.03961181640625,
0.03375244140625,
0.0267181396484375,
-0.016937255859375,
-0.039886474609375,
-0.0716552734375,
-0.00811767578125,
-0.01519775390625,
-0.0010814666748046875,
0.0110015869140625,
0.0276336669921875,
-0.0247344970703125,
0.05712890625,
-0.035369873046875,
-0.029571533203125,
-0.0401611328125,
0.0130157470703125,
0.018768310546875,
0.04632568359375,
0.0567626953125,
-0.0789794921875,
-0.059600830078125,
-0.0034160614013671875,
-0.055938720703125,
-0.010894775390625,
-0.01056671142578125,
-0.0242462158203125,
0.01548004150390625,
0.022369384765625,
-0.080322265625,
0.04022216796875,
0.049468994140625,
-0.010101318359375,
0.0426025390625,
-0.00856781005859375,
-0.004268646240234375,
-0.09637451171875,
-0.0017547607421875,
0.006626129150390625,
-0.0164642333984375,
-0.052093505859375,
0.043609619140625,
0.03497314453125,
-0.00968170166015625,
-0.0245819091796875,
0.0545654296875,
-0.055877685546875,
0.0007615089416503906,
-0.018524169921875,
0.02197265625,
0.016998291015625,
0.03948974609375,
-0.0031299591064453125,
0.051910400390625,
0.020477294921875,
-0.0186920166015625,
0.01332855224609375,
0.052490234375,
-0.0098114013671875,
0.0499267578125,
-0.055389404296875,
0.016204833984375,
-0.0009360313415527344,
0.02301025390625,
-0.05670166015625,
-0.01812744140625,
0.01910400390625,
-0.0423583984375,
0.050140380859375,
-0.04022216796875,
-0.048095703125,
-0.032684326171875,
-0.019744873046875,
0.0216522216796875,
0.046356201171875,
-0.038421630859375,
0.064697265625,
0.01800537109375,
-0.01070404052734375,
-0.0228118896484375,
-0.06597900390625,
-0.02099609375,
-0.01183319091796875,
-0.0819091796875,
0.046295166015625,
-0.032440185546875,
-0.0088043212890625,
-0.004291534423828125,
-0.00574493408203125,
-0.01241302490234375,
-0.019317626953125,
0.0275115966796875,
0.02899169921875,
-0.01812744140625,
0.005340576171875,
-0.006824493408203125,
-0.031463623046875,
-0.00583648681640625,
-0.0046539306640625,
0.038909912109375,
-0.026885986328125,
-0.0297698974609375,
-0.03924560546875,
0.0218658447265625,
0.035919189453125,
-0.0274658203125,
0.048736572265625,
0.06927490234375,
-0.02685546875,
-0.005786895751953125,
-0.033538818359375,
0.0004146099090576172,
-0.037445068359375,
0.0021457672119140625,
-0.0516357421875,
-0.0556640625,
0.052032470703125,
0.01088714599609375,
0.0034580230712890625,
0.049530029296875,
0.046966552734375,
-0.0006389617919921875,
0.05560302734375,
0.0887451171875,
-0.01348876953125,
0.057220458984375,
-0.031219482421875,
0.0254364013671875,
-0.0723876953125,
-0.0021648406982421875,
-0.04266357421875,
-0.0134429931640625,
-0.0635986328125,
-0.0310516357421875,
0.0304718017578125,
0.0193328857421875,
-0.00640106201171875,
0.04180908203125,
-0.0540771484375,
0.021697998046875,
0.042236328125,
-0.01226043701171875,
0.0221710205078125,
-0.006771087646484375,
0.0025310516357421875,
-0.0118865966796875,
-0.032196044921875,
-0.031494140625,
0.04608154296875,
0.02740478515625,
0.055938720703125,
0.030426025390625,
0.068359375,
-0.01226806640625,
0.02789306640625,
-0.06878662109375,
0.0282135009765625,
-0.00583648681640625,
-0.0386962890625,
-0.01108551025390625,
-0.0233917236328125,
-0.061187744140625,
0.01413726806640625,
0.0012369155883789062,
-0.0682373046875,
0.0014619827270507812,
0.005001068115234375,
-0.010986328125,
0.024871826171875,
-0.0645751953125,
0.06927490234375,
-0.0084686279296875,
-0.01050567626953125,
0.016204833984375,
-0.03533935546875,
0.037445068359375,
0.007293701171875,
0.01568603515625,
0.0140533447265625,
0.0203399658203125,
0.048095703125,
-0.046539306640625,
0.049530029296875,
-0.006023406982421875,
-0.006473541259765625,
0.024261474609375,
0.01617431640625,
0.046722412109375,
0.0094146728515625,
0.005218505859375,
0.007122039794921875,
0.00951385498046875,
-0.0285186767578125,
-0.059661865234375,
0.05828857421875,
-0.06396484375,
-0.04290771484375,
-0.021209716796875,
-0.049560546875,
0.0048370361328125,
0.031494140625,
0.033355712890625,
0.0152435302734375,
-0.0150909423828125,
0.003482818603515625,
0.05133056640625,
-0.01438140869140625,
0.0099029541015625,
0.02996826171875,
-0.034393310546875,
-0.007781982421875,
0.060516357421875,
0.006195068359375,
0.0086822509765625,
0.045623779296875,
0.0223388671875,
-0.00012695789337158203,
-0.021697998046875,
-0.050140380859375,
0.0304718017578125,
-0.04266357421875,
-0.0187530517578125,
-0.07086181640625,
-0.049407958984375,
-0.0350341796875,
-0.01371002197265625,
-0.0254364013671875,
-0.0382080078125,
-0.0106964111328125,
0.002521514892578125,
0.035675048828125,
0.052276611328125,
-0.0048980712890625,
0.033294677734375,
-0.061279296875,
0.037139892578125,
0.0194549560546875,
0.031158447265625,
-0.01192474365234375,
-0.0513916015625,
-0.021759033203125,
-0.0122833251953125,
-0.039154052734375,
-0.06439208984375,
0.02325439453125,
-0.004360198974609375,
0.04022216796875,
0.01806640625,
-0.01517486572265625,
0.04315185546875,
-0.03485107421875,
0.044097900390625,
0.014007568359375,
-0.065673828125,
0.0537109375,
-0.03466796875,
0.0288848876953125,
0.0151519775390625,
0.023468017578125,
-0.041046142578125,
-0.01012420654296875,
-0.06488037109375,
-0.058197021484375,
0.0582275390625,
0.02392578125,
0.0159912109375,
0.01459503173828125,
0.0153656005859375,
-0.004207611083984375,
-0.005466461181640625,
-0.076904296875,
-0.0286102294921875,
-0.019683837890625,
-0.0211944580078125,
0.0262603759765625,
-0.0298614501953125,
-0.01512908935546875,
-0.0244293212890625,
0.045135498046875,
-0.010589599609375,
0.046844482421875,
0.00504302978515625,
-0.0256500244140625,
-0.0106353759765625,
0.0172271728515625,
0.03741455078125,
0.01824951171875,
-0.0211334228515625,
-0.0221099853515625,
-0.0002694129943847656,
-0.03997802734375,
-0.015838623046875,
0.0286102294921875,
-0.025848388671875,
0.00685882568359375,
0.03570556640625,
0.0509033203125,
0.01158905029296875,
-0.017242431640625,
0.051239013671875,
-0.01393890380859375,
-0.0193023681640625,
-0.0313720703125,
-0.0015354156494140625,
0.0259246826171875,
0.0250396728515625,
0.0006399154663085938,
0.0006213188171386719,
0.008026123046875,
-0.0267486572265625,
0.01073455810546875,
0.030242919921875,
-0.025238037109375,
-0.03472900390625,
0.04864501953125,
-0.0026340484619140625,
-0.0380859375,
0.03619384765625,
-0.02960205078125,
-0.035552978515625,
0.039947509765625,
0.04791259765625,
0.06817626953125,
-0.00821685791015625,
0.006275177001953125,
0.0267333984375,
0.0296478271484375,
0.014495849609375,
0.02728271484375,
-0.01395416259765625,
-0.055267333984375,
-0.0162353515625,
-0.0634765625,
-0.0210723876953125,
0.01332855224609375,
-0.053070068359375,
0.023681640625,
-0.05029296875,
-0.006107330322265625,
0.01383209228515625,
0.014984130859375,
-0.0679931640625,
0.00591278076171875,
-0.00028228759765625,
0.08636474609375,
-0.04827880859375,
0.07586669921875,
0.0670166015625,
-0.06158447265625,
-0.0718994140625,
0.01129913330078125,
0.02642822265625,
-0.04986572265625,
0.049835205078125,
-0.005710601806640625,
-0.00130462646484375,
-0.014312744140625,
-0.040496826171875,
-0.06976318359375,
0.07501220703125,
0.01381683349609375,
-0.02545166015625,
-0.01297760009765625,
-0.022247314453125,
0.045135498046875,
-0.021484375,
0.0308685302734375,
0.019500732421875,
0.039031982421875,
-0.0030975341796875,
-0.0789794921875,
-0.0005207061767578125,
-0.042022705078125,
0.00409698486328125,
0.017852783203125,
-0.08209228515625,
0.0787353515625,
-0.010894775390625,
-0.0182037353515625,
0.033447265625,
0.06414794921875,
0.01116943359375,
0.00891876220703125,
0.0288238525390625,
0.0418701171875,
0.051605224609375,
-0.0176544189453125,
0.08740234375,
-0.028961181640625,
0.03533935546875,
0.08782958984375,
0.007358551025390625,
0.041717529296875,
0.018524169921875,
-0.0285797119140625,
0.065673828125,
0.0153961181640625,
-0.004833221435546875,
0.020111083984375,
-0.0004978179931640625,
-0.002552032470703125,
0.007411956787109375,
0.003437042236328125,
-0.0267333984375,
0.0276031494140625,
0.0298309326171875,
-0.035491943359375,
-0.00782012939453125,
-0.0015163421630859375,
0.013641357421875,
0.0067291259765625,
0.0053558349609375,
0.055389404296875,
0.01364898681640625,
-0.0152435302734375,
0.026580810546875,
0.00969696044921875,
0.057769775390625,
-0.051971435546875,
0.005573272705078125,
-0.0310821533203125,
0.0091094970703125,
-0.0291900634765625,
-0.050262451171875,
0.029632568359375,
0.00299072265625,
-0.041748046875,
-0.02789306640625,
0.06475830078125,
-0.021820068359375,
-0.0460205078125,
0.0228729248046875,
0.054046630859375,
0.01788330078125,
0.024444580078125,
-0.06988525390625,
0.019134521484375,
0.01251983642578125,
-0.037628173828125,
0.040008544921875,
0.04302978515625,
-0.01513671875,
0.049346923828125,
0.042724609375,
0.00458526611328125,
-0.007556915283203125,
0.0170135498046875,
0.06976318359375,
-0.0433349609375,
-0.039459228515625,
-0.040985107421875,
0.050811767578125,
-0.0007090568542480469,
-0.01611328125,
0.0654296875,
0.0338134765625,
0.06500244140625,
-0.006656646728515625,
0.061767578125,
-0.0294647216796875,
0.0438232421875,
-0.02825927734375,
0.06884765625,
-0.051483154296875,
-0.0021533966064453125,
-0.038238525390625,
-0.06317138671875,
-0.0196380615234375,
0.06866455078125,
-0.0199737548828125,
0.0245208740234375,
0.04791259765625,
0.0511474609375,
-0.0240478515625,
-0.0127716064453125,
0.0195465087890625,
0.02777099609375,
0.03082275390625,
0.0190277099609375,
0.026947021484375,
-0.051513671875,
0.04351806640625,
-0.03094482421875,
-0.0256500244140625,
-0.01433563232421875,
-0.0489501953125,
-0.06658935546875,
-0.055267333984375,
-0.026153564453125,
-0.0303955078125,
-0.0101165771484375,
0.051788330078125,
0.057769775390625,
-0.0477294921875,
-0.0015344619750976562,
0.01357269287109375,
0.022613525390625,
-0.0011997222900390625,
-0.02215576171875,
0.050994873046875,
-0.01421356201171875,
-0.0697021484375,
0.0161895751953125,
0.0164031982421875,
0.0257110595703125,
-0.02093505859375,
-0.006778717041015625,
0.0008482933044433594,
-0.007129669189453125,
0.039276123046875,
0.0341796875,
-0.0489501953125,
-0.01105499267578125,
-0.00913238525390625,
-0.026824951171875,
0.013641357421875,
0.041534423828125,
-0.041168212890625,
0.0170135498046875,
0.04595947265625,
0.027496337890625,
0.0435791015625,
-0.001399993896484375,
0.0245513916015625,
-0.06134033203125,
0.034454345703125,
-0.0120391845703125,
0.04644775390625,
0.0362548828125,
-0.030731201171875,
0.037445068359375,
0.006626129150390625,
-0.04205322265625,
-0.05035400390625,
0.02850341796875,
-0.0921630859375,
0.0007758140563964844,
0.0682373046875,
-0.0173187255859375,
-0.034423828125,
0.0287017822265625,
-0.0262451171875,
0.024749755859375,
-0.0243377685546875,
0.045166015625,
0.039947509765625,
-0.00824737548828125,
-0.041656494140625,
-0.0484619140625,
0.04473876953125,
0.003910064697265625,
-0.05242919921875,
-0.0214691162109375,
0.0157928466796875,
0.02276611328125,
0.052093505859375,
0.060455322265625,
-0.014678955078125,
0.01525115966796875,
0.003917694091796875,
0.034881591796875,
0.004749298095703125,
-0.0163421630859375,
-0.01383209228515625,
-0.0027980804443359375,
0.00948333740234375,
0.005413055419921875
]
] |
stabilityai/stablelm-base-alpha-3b | 2023-10-19T04:58:32.000Z | [
"transformers",
"pytorch",
"gpt_neox",
"text-generation",
"causal-lm",
"en",
"license:cc-by-sa-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | stabilityai | null | null | stabilityai/stablelm-base-alpha-3b | 82 | 7,458 | transformers | 2023-04-17T22:14:52 | ---
language:
- en
license:
- cc-by-sa-4.0
tags:
- causal-lm
---
# StableLM-Base-Alpha
📢 **DISCLAIMER**: The StableLM-Base-Alpha models have been superseded. Find the latest versions in the Stable LM Collection [here](https://huggingface.co/collections/stabilityai/stable-lm-650852cfd55dd4e15cdcb30a).
## Model Description
`StableLM-Base-Alpha` is a suite of 3B and 7B parameter decoder-only language models pre-trained on a diverse collection of English and Code datasets with a sequence length of 4096 to push beyond the context window limitations of existing open-source language models.
## Usage
Get started generating text with `StableLM-Base-Alpha` by using the following code snippet:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("StabilityAI/stablelm-base-alpha-3b")
model = AutoModelForCausalLM.from_pretrained("StabilityAI/stablelm-base-alpha-3b")
model.half().cuda()
inputs = tokenizer("What's your mood today?", return_tensors="pt").to("cuda")
tokens = model.generate(
**inputs,
max_new_tokens=64,
temperature=0.7,
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 models are auto-regressive language models based on the NeoX transformer architecture.
* **Language(s)**: English
* **Library**: [GPT-NeoX](https://github.com/EleutherAI/gpt-neox)
* **License**: Base model checkpoints (StableLM-Base-Alpha) are licensed under the Creative Commons license (CC BY-SA-4.0). Under the license, you must give credit to Stability AI, provide a link to the license, and indicate if changes were made. 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`
## Training
| Parameters | Hidden Size | Layers | Heads | Sequence Length |
|------------|-------------|--------|-------|-----------------|
| 3B | 4096 | 16 | 32 | 4096 |
| 7B | 6144 | 16 | 48 | 4096 |
### Training Dataset
`StableLM-Base-Alpha` is pre-trained on a new experimental dataset built atop [The Pile](https://huggingface.co/datasets/EleutherAI/the_pile) and is threes times larger at approximately 1.5T tokens.
### Training Procedure
Models are pre-trained on the aforementioned dataset in mixed-precision (FP16), optimized with Adam, 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](https://github.com/Stability-AI/StableLM/blob/main/configs/stablelm-base-alpha-3b.yaml).
## 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 contain offensive or inappropriate content even after applying data cleansing filters which can be reflected in generated text. We recommend users exercise reasonable 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.
## Citations
```bibtext
@software{gpt-neox-library,
title = {{GPT-NeoX: Large Scale Autoregressive Language Modeling in PyTorch}},
author = {Andonian, Alex and Anthony, Quentin and Biderman, Stella and Black, Sid and Gali, Preetham and Gao, Leo and Hallahan, Eric and Levy-Kramer, Josh and Leahy, Connor and Nestler, Lucas and Parker, Kip and Pieler, Michael and Purohit, Shivanshu and Songz, Tri and Phil, Wang and Weinbach, Samuel},
url = {https://www.github.com/eleutherai/gpt-neox},
doi = {10.5281/zenodo.5879544},
month = {8},
year = {2021},
version = {0.0.1},
}
``` | 3,979 | [
[
-0.0177001953125,
-0.06414794921875,
0.0083160400390625,
0.01218414306640625,
-0.023040771484375,
-0.01145172119140625,
-0.0231170654296875,
-0.03717041015625,
0.007343292236328125,
0.0181732177734375,
-0.0247039794921875,
-0.042755126953125,
-0.04400634765625,
-0.0029735565185546875,
-0.0304412841796875,
0.0889892578125,
-0.0006160736083984375,
-0.0116119384765625,
0.004154205322265625,
-0.01953125,
-0.019378662109375,
-0.056610107421875,
-0.049896240234375,
-0.012786865234375,
0.021484375,
-0.0016431808471679688,
0.072509765625,
0.069091796875,
0.0232391357421875,
0.0243988037109375,
-0.01523590087890625,
-0.016571044921875,
-0.033843994140625,
0.0181732177734375,
0.026123046875,
-0.008758544921875,
-0.0557861328125,
-0.0010852813720703125,
0.06097412109375,
0.027008056640625,
-0.027923583984375,
0.0156707763671875,
-0.002109527587890625,
0.0218658447265625,
-0.044525146484375,
0.018585205078125,
-0.041778564453125,
-0.027984619140625,
-0.0027103424072265625,
0.0308380126953125,
-0.02447509765625,
-0.0193023681640625,
-0.0035610198974609375,
-0.034912109375,
-0.00408172607421875,
-0.0027599334716796875,
0.10003662109375,
0.024078369140625,
-0.01200103759765625,
0.00519561767578125,
-0.04473876953125,
0.054779052734375,
-0.078857421875,
0.03717041015625,
0.036163330078125,
0.0014438629150390625,
0.0034656524658203125,
-0.052734375,
-0.0299072265625,
-0.025146484375,
0.0017957687377929688,
-0.00007730722427368164,
-0.0182952880859375,
0.004245758056640625,
0.021148681640625,
0.01454925537109375,
-0.057464599609375,
0.007518768310546875,
-0.014007568359375,
-0.03302001953125,
0.037353515625,
0.02142333984375,
0.007686614990234375,
-0.003429412841796875,
-0.0173797607421875,
-0.021759033203125,
-0.035980224609375,
-0.0005970001220703125,
0.01424407958984375,
0.02947998046875,
-0.03680419921875,
0.022613525390625,
0.00826263427734375,
0.053070068359375,
0.006610870361328125,
-0.0107421875,
0.048614501953125,
-0.0308380126953125,
-0.022613525390625,
-0.01505279541015625,
0.09393310546875,
0.0157623291015625,
-0.0025081634521484375,
-0.0021533966064453125,
-0.033203125,
0.0164794921875,
-0.005336761474609375,
-0.0699462890625,
-0.011260986328125,
0.021026611328125,
-0.0277557373046875,
-0.0194854736328125,
-0.0032100677490234375,
-0.044830322265625,
-0.0031108856201171875,
-0.01465606689453125,
0.03741455078125,
-0.03521728515625,
-0.039398193359375,
0.0082244873046875,
0.0092620849609375,
0.0219268798828125,
0.0017852783203125,
-0.05450439453125,
0.03033447265625,
0.03106689453125,
0.056304931640625,
-0.006603240966796875,
-0.035797119140625,
-0.035736083984375,
-0.007770538330078125,
-0.0177001953125,
0.01486968994140625,
-0.022491455078125,
-0.014556884765625,
-0.0033416748046875,
0.011260986328125,
-0.0034389495849609375,
-0.0178070068359375,
0.0192108154296875,
-0.030242919921875,
0.0249481201171875,
0.017486572265625,
-0.017303466796875,
-0.0003426074981689453,
0.037017822265625,
-0.035400390625,
0.09075927734375,
0.023284912109375,
-0.057586669921875,
0.007747650146484375,
-0.0217742919921875,
-0.020904541015625,
-0.0170135498046875,
-0.003658294677734375,
-0.059295654296875,
-0.0229644775390625,
0.00284576416015625,
0.0132293701171875,
-0.0220184326171875,
0.03265380859375,
-0.0260467529296875,
-0.01282501220703125,
-0.00377655029296875,
-0.0311126708984375,
0.07421875,
0.0157928466796875,
-0.05230712890625,
0.0255279541015625,
-0.0609130859375,
0.00014162063598632812,
0.01200103759765625,
-0.0181121826171875,
-0.0121917724609375,
-0.0181121826171875,
0.007061004638671875,
0.0276336669921875,
0.038726806640625,
-0.018463134765625,
0.0134735107421875,
-0.02984619140625,
0.033447265625,
0.04681396484375,
-0.01465606689453125,
0.0244140625,
-0.0088043212890625,
0.049774169921875,
0.00948333740234375,
0.0273284912109375,
-0.006137847900390625,
-0.04217529296875,
-0.056060791015625,
-0.02044677734375,
0.0271759033203125,
0.051788330078125,
-0.03729248046875,
0.04449462890625,
-0.0160064697265625,
-0.038787841796875,
-0.028472900390625,
0.01465606689453125,
0.047332763671875,
0.04315185546875,
0.035675048828125,
-0.0080108642578125,
-0.051116943359375,
-0.05645751953125,
0.0219573974609375,
-0.039947509765625,
0.0305023193359375,
0.000042557716369628906,
0.034271240234375,
-0.04443359375,
0.062042236328125,
-0.0181732177734375,
0.004924774169921875,
-0.005710601806640625,
0.0174560546875,
0.0333251953125,
0.046844482421875,
0.05645751953125,
-0.033538818359375,
-0.0352783203125,
-0.00855255126953125,
-0.048919677734375,
0.00504302978515625,
0.01433563232421875,
-0.014923095703125,
0.042144775390625,
0.0283203125,
-0.0665283203125,
0.01953125,
0.0560302734375,
-0.044281005859375,
0.04522705078125,
-0.0177459716796875,
-0.02545166015625,
-0.09783935546875,
0.02069091796875,
0.0029964447021484375,
-0.022247314453125,
-0.039703369140625,
-0.005420684814453125,
0.00927734375,
-0.007175445556640625,
-0.03912353515625,
0.0521240234375,
-0.041717529296875,
0.007450103759765625,
-0.014984130859375,
-0.0004489421844482422,
-0.002197265625,
0.027130126953125,
0.00396728515625,
0.044219970703125,
0.0733642578125,
-0.041778564453125,
0.007633209228515625,
0.011383056640625,
0.0124969482421875,
-0.01131439208984375,
-0.05853271484375,
0.01490020751953125,
0.0015916824340820312,
0.00971221923828125,
-0.056640625,
0.01013946533203125,
0.03900146484375,
-0.041473388671875,
0.0361328125,
-0.0256805419921875,
-0.027923583984375,
-0.0281829833984375,
-0.01308441162109375,
0.032745361328125,
0.059173583984375,
-0.00843048095703125,
0.047607421875,
0.034088134765625,
-0.008026123046875,
-0.0738525390625,
-0.045867919921875,
-0.0088043212890625,
-0.0175933837890625,
-0.035614013671875,
0.0030879974365234375,
-0.0184326171875,
-0.0255279541015625,
0.00939178466796875,
-0.0006055831909179688,
0.0013017654418945312,
0.00914764404296875,
0.0250701904296875,
0.042083740234375,
-0.026031494140625,
-0.01306915283203125,
-0.0181732177734375,
-0.0217132568359375,
0.01312255859375,
-0.0303955078125,
0.059051513671875,
-0.05157470703125,
0.01346588134765625,
-0.0330810546875,
0.00876617431640625,
0.06890869140625,
-0.020294189453125,
0.07611083984375,
0.058013916015625,
-0.03167724609375,
0.01340484619140625,
-0.0246734619140625,
-0.0287322998046875,
-0.03289794921875,
0.03472900390625,
-0.00620269775390625,
-0.053558349609375,
0.0657958984375,
0.043426513671875,
0.0199127197265625,
0.05682373046875,
0.054290771484375,
0.019500732421875,
0.09173583984375,
0.042266845703125,
-0.02984619140625,
0.0350341796875,
-0.047821044921875,
-0.0081634521484375,
-0.048370361328125,
0.0005393028259277344,
-0.0443115234375,
-0.0007619857788085938,
-0.041778564453125,
-0.0219268798828125,
-0.0027027130126953125,
0.0054168701171875,
-0.061065673828125,
0.031890869140625,
-0.036163330078125,
0.002536773681640625,
0.0291900634765625,
-0.0150299072265625,
-0.0031681060791015625,
-0.01000213623046875,
-0.0184326171875,
0.015411376953125,
-0.0499267578125,
-0.0270233154296875,
0.07220458984375,
0.042755126953125,
0.06768798828125,
-0.00022494792938232422,
0.047332763671875,
-0.0035190582275390625,
0.02838134765625,
-0.047210693359375,
0.037811279296875,
-0.01288604736328125,
-0.051727294921875,
-0.0258026123046875,
-0.047821044921875,
-0.07977294921875,
0.005222320556640625,
-0.0201416015625,
-0.037628173828125,
0.0247650146484375,
0.025482177734375,
-0.02960205078125,
0.01165771484375,
-0.0404052734375,
0.07574462890625,
-0.040924072265625,
-0.032684326171875,
0.005664825439453125,
-0.06390380859375,
0.01296234130859375,
0.01263427734375,
0.0211944580078125,
-0.0113677978515625,
-0.01320648193359375,
0.049163818359375,
-0.039520263671875,
0.06829833984375,
-0.0267181396484375,
-0.0018262863159179688,
0.020904541015625,
0.004367828369140625,
0.04443359375,
0.0140380859375,
-0.0267333984375,
0.031982421875,
-0.00537109375,
-0.02655029296875,
-0.0231475830078125,
0.052337646484375,
-0.10003662109375,
-0.033721923828125,
-0.04852294921875,
-0.03704833984375,
0.00194549560546875,
0.04119873046875,
0.0174560546875,
0.037017822265625,
0.00489044189453125,
0.020263671875,
0.02899169921875,
0.00760650634765625,
0.045989990234375,
0.044403076171875,
-0.0267791748046875,
-0.059234619140625,
0.059356689453125,
0.00992584228515625,
0.01505279541015625,
0.0011749267578125,
0.020782470703125,
-0.03765869140625,
-0.061798095703125,
-0.038787841796875,
0.0297393798828125,
-0.0452880859375,
-0.0276336669921875,
-0.041290283203125,
-0.0203857421875,
-0.03460693359375,
0.01090240478515625,
-0.043426513671875,
-0.0287322998046875,
-0.0270538330078125,
-0.00865936279296875,
0.038726806640625,
0.0301666259765625,
0.0008025169372558594,
0.01953125,
-0.056060791015625,
0.018157958984375,
0.0094451904296875,
0.024658203125,
-0.020172119140625,
-0.053070068359375,
-0.0323486328125,
0.0173797607421875,
-0.006237030029296875,
-0.050384521484375,
0.0509033203125,
0.017547607421875,
0.05322265625,
0.0265350341796875,
0.01532745361328125,
0.04400634765625,
-0.0189056396484375,
0.06427001953125,
0.01314544677734375,
-0.06268310546875,
0.042694091796875,
-0.03814697265625,
0.02252197265625,
0.0467529296875,
0.0300140380859375,
-0.01039886474609375,
-0.04473876953125,
-0.063232421875,
-0.0933837890625,
0.054290771484375,
0.015350341796875,
0.01189422607421875,
-0.01335906982421875,
0.04510498046875,
-0.007781982421875,
0.0085296630859375,
-0.07916259765625,
-0.0355224609375,
-0.041961669921875,
-0.026519775390625,
-0.01291656494140625,
-0.00653839111328125,
-0.00959014892578125,
-0.028289794921875,
0.070556640625,
-0.007843017578125,
0.0181121826171875,
0.0056915283203125,
-0.016143798828125,
-0.0172576904296875,
-0.006000518798828125,
0.04254150390625,
0.051666259765625,
-0.03961181640625,
0.0075836181640625,
0.006214141845703125,
-0.0579833984375,
0.0168609619140625,
0.03106689453125,
-0.034515380859375,
-0.004180908203125,
0.003307342529296875,
0.0870361328125,
-0.0019702911376953125,
-0.0306549072265625,
0.0218353271484375,
-0.0265350341796875,
-0.024627685546875,
-0.02947998046875,
0.0033321380615234375,
0.0011997222900390625,
-0.008087158203125,
0.019561767578125,
0.00909423828125,
-0.0222015380859375,
-0.033935546875,
0.0161285400390625,
0.03326416015625,
-0.031951904296875,
-0.032073974609375,
0.05670166015625,
0.0098114013671875,
-0.0189361572265625,
0.070556640625,
-0.01013946533203125,
-0.027252197265625,
0.046234130859375,
0.05865478515625,
0.07208251953125,
-0.0189056396484375,
0.0010089874267578125,
0.04302978515625,
0.033294677734375,
-0.0207366943359375,
0.022430419921875,
0.027496337890625,
-0.0633544921875,
-0.027435302734375,
-0.049163818359375,
-0.0201568603515625,
0.0278167724609375,
-0.047760009765625,
0.028106689453125,
-0.055572509765625,
-0.04071044921875,
-0.03472900390625,
0.007720947265625,
-0.021636962890625,
0.0208282470703125,
0.022216796875,
0.05987548828125,
-0.06854248046875,
0.0762939453125,
0.07281494140625,
-0.04205322265625,
-0.07635498046875,
0.0001558065414428711,
-0.01287078857421875,
-0.044677734375,
0.0266571044921875,
0.00799560546875,
-0.015411376953125,
0.0168609619140625,
-0.036468505859375,
-0.0784912109375,
0.0810546875,
0.050537109375,
-0.03741455078125,
-0.00439453125,
-0.0095367431640625,
0.043731689453125,
-0.0239715576171875,
0.035369873046875,
0.0258636474609375,
0.036346435546875,
-0.00862884521484375,
-0.06292724609375,
0.01049041748046875,
-0.052764892578125,
-0.0074462890625,
0.0221710205078125,
-0.058929443359375,
0.08270263671875,
0.0002453327178955078,
0.006908416748046875,
-0.0005965232849121094,
0.052978515625,
0.042572021484375,
0.004932403564453125,
0.045379638671875,
0.06317138671875,
0.04010009765625,
-0.0106353759765625,
0.0704345703125,
-0.062164306640625,
0.04351806640625,
0.07147216796875,
-0.0015668869018554688,
0.07244873046875,
0.0208740234375,
-0.0018291473388671875,
0.0518798828125,
0.045867919921875,
-0.00370025634765625,
0.0177001953125,
-0.01302337646484375,
0.006237030029296875,
-0.0253753662109375,
0.0196685791015625,
-0.051361083984375,
0.0125732421875,
0.03021240234375,
-0.02850341796875,
-0.0015201568603515625,
-0.008636474609375,
0.0201416015625,
-0.0242919921875,
-0.0167388916015625,
0.041290283203125,
0.01007080078125,
-0.04425048828125,
0.09832763671875,
-0.002887725830078125,
0.046966552734375,
-0.0552978515625,
0.018829345703125,
-0.0178680419921875,
0.0244598388671875,
0.0023517608642578125,
-0.042205810546875,
0.0188140869140625,
-0.003993988037109375,
-0.018463134765625,
-0.017303466796875,
0.043670654296875,
-0.0266571044921875,
-0.03936767578125,
0.038360595703125,
0.02899169921875,
0.0013914108276367188,
0.01372528076171875,
-0.0830078125,
0.0252838134765625,
-0.0201416015625,
-0.040313720703125,
0.02471923828125,
0.0186920166015625,
-0.0097503662109375,
0.04473876953125,
0.043670654296875,
0.0023593902587890625,
0.004863739013671875,
0.0133056640625,
0.07421875,
-0.049896240234375,
-0.037811279296875,
-0.062225341796875,
0.0548095703125,
0.0128936767578125,
-0.037750244140625,
0.059356689453125,
0.049896240234375,
0.0364990234375,
0.007251739501953125,
0.049224853515625,
-0.02166748046875,
0.00995635986328125,
-0.0240020751953125,
0.057159423828125,
-0.039215087890625,
0.013336181640625,
-0.0217437744140625,
-0.07037353515625,
-0.0258941650390625,
0.055694580078125,
-0.015594482421875,
0.0196990966796875,
0.03704833984375,
0.0635986328125,
0.0075531005859375,
-0.034149169921875,
0.0100860595703125,
0.05743408203125,
0.0160980224609375,
0.029266357421875,
0.0577392578125,
-0.052276611328125,
0.056060791015625,
-0.03460693359375,
-0.015716552734375,
-0.018310546875,
-0.06500244140625,
-0.06298828125,
-0.0372314453125,
-0.03631591796875,
-0.06402587890625,
0.0208892822265625,
0.07598876953125,
0.06768798828125,
-0.06475830078125,
-0.0311737060546875,
-0.0207672119140625,
-0.004192352294921875,
-0.017669677734375,
-0.0142822265625,
0.0306854248046875,
-0.02490234375,
-0.040618896484375,
0.01763916015625,
-0.0010118484497070312,
0.0199127197265625,
-0.02508544921875,
-0.037353515625,
-0.0210723876953125,
-0.01360321044921875,
0.021636962890625,
0.04034423828125,
-0.039642333984375,
-0.00982666015625,
0.01849365234375,
-0.0180511474609375,
0.0161285400390625,
0.0235137939453125,
-0.0509033203125,
0.0087432861328125,
0.041656494140625,
0.024322509765625,
0.038848876953125,
0.0048828125,
0.038330078125,
-0.037109375,
0.035675048828125,
0.030303955078125,
0.038238525390625,
0.020660400390625,
-0.0190277099609375,
0.016876220703125,
0.0276336669921875,
-0.0467529296875,
-0.0667724609375,
-0.0008745193481445312,
-0.07598876953125,
-0.0005841255187988281,
0.10076904296875,
-0.015777587890625,
-0.036102294921875,
-0.01262664794921875,
-0.01047515869140625,
0.0259246826171875,
-0.0428466796875,
0.04254150390625,
0.0335693359375,
-0.0012607574462890625,
-0.040802001953125,
-0.0187835693359375,
0.03704833984375,
0.02752685546875,
-0.04376220703125,
0.015838623046875,
0.0411376953125,
0.02764892578125,
0.026580810546875,
0.042205810546875,
-0.02276611328125,
0.023223876953125,
-0.00617218017578125,
0.0231475830078125,
-0.019317626953125,
-0.0210418701171875,
-0.03765869140625,
0.00344085693359375,
0.019683837890625,
0.003993988037109375
]
] |
winglian/basilisk-4b | 2023-09-25T22:16:55.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:Open-Orca/OpenOrca",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | winglian | null | null | winglian/basilisk-4b | 4 | 7,449 | transformers | 2023-09-21T11:41:09 | ---
datasets:
- Open-Orca/OpenOrca
library_name: transformers
tags:
- llama
---
# Basilisk 4B
[<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)
Built on `winglian/llama-2-4b`, a 4B parameter Llama-2 model, this model is finetuned with open orca CoT data.
```
hf-causal-experimental (pretrained=winglian/basilisk-4b,use_accelerate=True,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 0, batch_size: None
| Task |Version| Metric |Value | |Stderr|
|------------------------------------------------|------:|---------------------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.2362|_ |0.0267|
| | |acc_norm |0.2283|_ |0.0264|
|agieval_logiqa_en | 0|acc |0.2688|_ |0.0174|
| | |acc_norm |0.2811|_ |0.0176|
|agieval_lsat_ar | 0|acc |0.2130|_ |0.0271|
| | |acc_norm |0.1913|_ |0.0260|
|agieval_lsat_lr | 0|acc |0.2255|_ |0.0185|
| | |acc_norm |0.2745|_ |0.0198|
|agieval_lsat_rc | 0|acc |0.2305|_ |0.0257|
| | |acc_norm |0.2491|_ |0.0264|
|agieval_sat_en | 0|acc |0.3641|_ |0.0336|
| | |acc_norm |0.3495|_ |0.0333|
|agieval_sat_en_without_passage | 0|acc |0.2427|_ |0.0299|
| | |acc_norm |0.2427|_ |0.0299|
|agieval_sat_math | 0|acc |0.2318|_ |0.0285|
| | |acc_norm |0.2091|_ |0.0275|
|bigbench_causal_judgement | 0|multiple_choice_grade|0.5000|_ |0.0364|
|bigbench_date_understanding | 0|multiple_choice_grade|0.3930|_ |0.0255|
|bigbench_disambiguation_qa | 0|multiple_choice_grade|0.2674|_ |0.0276|
|bigbench_geometric_shapes | 0|multiple_choice_grade|0.1838|_ |0.0205|
| | |exact_str_match |0.0279|_ |0.0087|
|bigbench_logical_deduction_five_objects | 0|multiple_choice_grade|0.2380|_ |0.0191|
|bigbench_logical_deduction_seven_objects | 0|multiple_choice_grade|0.1843|_ |0.0147|
|bigbench_logical_deduction_three_objects | 0|multiple_choice_grade|0.3800|_ |0.0281|
|bigbench_movie_recommendation | 0|multiple_choice_grade|0.3480|_ |0.0213|
|bigbench_navigate | 0|multiple_choice_grade|0.5000|_ |0.0158|
|bigbench_reasoning_about_colored_objects | 0|multiple_choice_grade|0.3680|_ |0.0108|
|bigbench_ruin_names | 0|multiple_choice_grade|0.2746|_ |0.0211|
|bigbench_salient_translation_error_detection | 0|multiple_choice_grade|0.2806|_ |0.0142|
|bigbench_snarks | 0|multiple_choice_grade|0.4972|_ |0.0373|
|bigbench_sports_understanding | 0|multiple_choice_grade|0.4939|_ |0.0159|
|bigbench_temporal_sequences | 0|multiple_choice_grade|0.2740|_ |0.0141|
|bigbench_tracking_shuffled_objects_five_objects | 0|multiple_choice_grade|0.1904|_ |0.0111|
|bigbench_tracking_shuffled_objects_seven_objects| 0|multiple_choice_grade|0.1394|_ |0.0083|
|bigbench_tracking_shuffled_objects_three_objects| 0|multiple_choice_grade|0.3800|_ |0.0281|
hf-causal-experimental (pretrained=winglian/basilisk-4b,use_accelerate=True,trust_remote_code=True), limit: None, provide_description: False, num_fewshot: 0, batch_size: 12
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.3285|_ |0.0137|
| | |acc_norm|0.3532|_ |0.0140|
|arc_easy | 0|acc |0.6364|_ |0.0099|
| | |acc_norm|0.6035|_ |0.0100|
|boolq | 1|acc |0.7196|_ |0.0079|
|hellaswag | 0|acc |0.4239|_ |0.0049|
| | |acc_norm|0.5473|_ |0.0050|
|openbookqa | 0|acc |0.2220|_ |0.0186|
| | |acc_norm|0.3320|_ |0.0211|
|piqa | 0|acc |0.6937|_ |0.0108|
| | |acc_norm|0.6921|_ |0.0108|
|winogrande | 0|acc |0.5399|_ |0.0140|
```
| 5,160 | [
[
-0.03900146484375,
-0.03485107421875,
0.025970458984375,
0.0148162841796875,
-0.01947021484375,
0.007415771484375,
0.002002716064453125,
-0.0244598388671875,
0.045928955078125,
0.0024356842041015625,
-0.050445556640625,
-0.046356201171875,
-0.05023193359375,
0.0035076141357421875,
0.00908660888671875,
0.06341552734375,
-0.0044097900390625,
-0.0005211830139160156,
0.0085906982421875,
-0.0179290771484375,
-0.0308074951171875,
-0.006443023681640625,
-0.05401611328125,
-0.0150604248046875,
0.029022216796875,
0.035430908203125,
0.045257568359375,
0.049560546875,
0.047210693359375,
0.01611328125,
-0.02093505859375,
0.01058197021484375,
-0.0214080810546875,
-0.023101806640625,
0.015167236328125,
-0.0256195068359375,
-0.051605224609375,
0.0139007568359375,
0.042083740234375,
0.044403076171875,
-0.0025234222412109375,
0.0406494140625,
0.00490570068359375,
0.049102783203125,
-0.0278167724609375,
0.0253448486328125,
-0.00305938720703125,
-0.0016317367553710938,
-0.01303863525390625,
-0.01166534423828125,
0.001361846923828125,
-0.036102294921875,
-0.0107421875,
-0.052154541015625,
0.00975799560546875,
0.01751708984375,
0.102783203125,
0.022216796875,
-0.0207061767578125,
-0.0163726806640625,
-0.01557159423828125,
0.0587158203125,
-0.06195068359375,
0.00335693359375,
0.04510498046875,
-0.0007762908935546875,
-0.00881195068359375,
-0.0374755859375,
-0.055450439453125,
0.01416015625,
-0.0243377685546875,
0.015777587890625,
-0.0113372802734375,
-0.00952911376953125,
0.023193359375,
0.036407470703125,
-0.0509033203125,
0.01125335693359375,
-0.05029296875,
-0.018310546875,
0.056976318359375,
0.039398193359375,
0.0085296630859375,
-0.0228729248046875,
-0.0301971435546875,
-0.02642822265625,
-0.032958984375,
0.04364013671875,
0.032196044921875,
0.01116943359375,
-0.04083251953125,
0.042724609375,
-0.0205841064453125,
0.038330078125,
0.020599365234375,
-0.0294342041015625,
0.0687255859375,
-0.032196044921875,
-0.017730712890625,
0.0019311904907226562,
0.06439208984375,
0.048187255859375,
-0.0169219970703125,
0.025146484375,
0.004764556884765625,
-0.000209808349609375,
-0.01061248779296875,
-0.06536865234375,
-0.009765625,
0.036865234375,
-0.03857421875,
-0.016143798828125,
0.01371002197265625,
-0.056793212890625,
-0.004772186279296875,
-0.017791748046875,
0.0229339599609375,
-0.03350830078125,
-0.012603759765625,
-0.006183624267578125,
-0.02252197265625,
0.04229736328125,
0.01369476318359375,
-0.060882568359375,
0.008758544921875,
0.02593994140625,
0.06195068359375,
-0.01224517822265625,
-0.01983642578125,
-0.01346588134765625,
0.0163726806640625,
-0.040008544921875,
0.053131103515625,
-0.01361846923828125,
-0.0250701904296875,
-0.0249786376953125,
0.0163421630859375,
-0.01476287841796875,
-0.02471923828125,
0.04736328125,
-0.0123748779296875,
0.00606536865234375,
-0.0538330078125,
-0.017242431640625,
-0.018280029296875,
0.0261688232421875,
-0.049652099609375,
0.095947265625,
0.00966644287109375,
-0.062103271484375,
0.0498046875,
-0.033782958984375,
-0.005428314208984375,
-0.0155029296875,
-0.0226287841796875,
-0.05548095703125,
-0.0242156982421875,
0.0269317626953125,
0.0217437744140625,
-0.0367431640625,
0.01451873779296875,
-0.0170745849609375,
-0.028228759765625,
-0.0004718303680419922,
-0.01605224609375,
0.09185791015625,
0.0144500732421875,
-0.042510986328125,
0.01082611083984375,
-0.07086181640625,
0.00669097900390625,
0.017181396484375,
-0.033721923828125,
0.000667572021484375,
-0.02215576171875,
-0.01447296142578125,
0.0180206298828125,
0.0245819091796875,
-0.038238525390625,
0.0194244384765625,
-0.0128173828125,
0.0217742919921875,
0.0684814453125,
0.01363372802734375,
0.011962890625,
-0.0465087890625,
0.0299072265625,
0.0226898193359375,
0.0187225341796875,
0.01549530029296875,
-0.04443359375,
-0.058349609375,
-0.05609130859375,
-0.003631591796875,
0.038421630859375,
-0.0204315185546875,
0.044097900390625,
-0.00951385498046875,
-0.050689697265625,
-0.03717041015625,
0.0004608631134033203,
0.034698486328125,
0.0406494140625,
0.0352783203125,
-0.01319122314453125,
-0.0333251953125,
-0.07562255859375,
0.003070831298828125,
-0.016510009765625,
0.01143646240234375,
0.0447998046875,
0.06884765625,
-0.01427459716796875,
0.054290771484375,
-0.06591796875,
-0.040252685546875,
-0.0151824951171875,
0.00595855712890625,
0.0557861328125,
0.04656982421875,
0.05169677734375,
-0.04071044921875,
-0.04766845703125,
-0.01088714599609375,
-0.05755615234375,
0.00001537799835205078,
-0.0038604736328125,
-0.0198516845703125,
0.01203155517578125,
0.0177459716796875,
-0.055908203125,
0.060882568359375,
0.027557373046875,
-0.050537109375,
0.061126708984375,
-0.0249481201171875,
0.0227203369140625,
-0.06878662109375,
0.0250701904296875,
-0.0161895751953125,
0.023406982421875,
-0.0282440185546875,
-0.0212249755859375,
0.0105743408203125,
0.0101165771484375,
-0.02618408203125,
0.049560546875,
-0.04901123046875,
-0.0036716461181640625,
0.0221405029296875,
-0.004238128662109375,
-0.007366180419921875,
0.050567626953125,
0.0017404556274414062,
0.0634765625,
0.057952880859375,
-0.03240966796875,
0.011474609375,
0.0222930908203125,
-0.034027099609375,
0.0355224609375,
-0.0447998046875,
-0.0121307373046875,
-0.00743865966796875,
0.0149688720703125,
-0.09600830078125,
-0.035736083984375,
0.021331787109375,
-0.037567138671875,
0.005931854248046875,
0.017303466796875,
-0.0161895751953125,
-0.05511474609375,
-0.052520751953125,
0.0240631103515625,
0.023681640625,
-0.02734375,
0.019134521484375,
0.01029205322265625,
-0.002712249755859375,
-0.05029296875,
-0.0518798828125,
-0.02093505859375,
-0.01065826416015625,
-0.041290283203125,
0.032257080078125,
-0.007320404052734375,
-0.007709503173828125,
0.004302978515625,
-0.0158538818359375,
-0.009185791015625,
-0.00347137451171875,
0.028778076171875,
0.035430908203125,
-0.01910400390625,
-0.0256805419921875,
0.006786346435546875,
-0.004360198974609375,
0.00897979736328125,
0.0154266357421875,
0.0396728515625,
-0.0087127685546875,
-0.02349853515625,
-0.0484619140625,
0.00826263427734375,
0.0455322265625,
-0.0212554931640625,
0.075439453125,
0.035797119140625,
-0.01568603515625,
0.00785064697265625,
-0.026580810546875,
-0.0033245086669921875,
-0.0343017578125,
0.00737762451171875,
-0.0284423828125,
-0.0496826171875,
0.059356689453125,
0.0239105224609375,
0.0108642578125,
0.0540771484375,
0.0343017578125,
-0.0012063980102539062,
0.0660400390625,
0.0166168212890625,
-0.003231048583984375,
0.0170440673828125,
-0.05145263671875,
0.0074615478515625,
-0.069580078125,
-0.04669189453125,
-0.04229736328125,
-0.0311279296875,
-0.034515380859375,
-0.023895263671875,
0.02496337890625,
0.006778717041015625,
-0.057403564453125,
0.02178955078125,
-0.04510498046875,
0.0133056640625,
0.055938720703125,
0.03125,
0.005352020263671875,
-0.0100860595703125,
-0.03753662109375,
-0.00591278076171875,
-0.0304412841796875,
-0.02197265625,
0.09478759765625,
0.00202178955078125,
0.041534423828125,
0.0299530029296875,
0.054168701171875,
0.0204010009765625,
0.01407623291015625,
-0.0252685546875,
0.0302886962890625,
0.018798828125,
-0.07147216796875,
-0.0244140625,
-0.01554107666015625,
-0.0714111328125,
0.035186767578125,
-0.01371002197265625,
-0.06707763671875,
0.039947509765625,
0.01078033447265625,
-0.02978515625,
0.021209716796875,
-0.040802001953125,
0.06512451171875,
-0.01708984375,
-0.0406494140625,
-0.0007834434509277344,
-0.04913330078125,
0.0297088623046875,
-0.0009512901306152344,
0.0295867919921875,
-0.0157012939453125,
-0.0036602020263671875,
0.07537841796875,
-0.05633544921875,
0.05145263671875,
-0.013885498046875,
0.01271820068359375,
0.037078857421875,
-0.01568603515625,
0.04119873046875,
0.0085906982421875,
-0.007160186767578125,
0.005489349365234375,
0.0000655055046081543,
-0.048980712890625,
-0.007640838623046875,
0.050445556640625,
-0.07440185546875,
-0.062103271484375,
-0.07177734375,
-0.034698486328125,
0.01226806640625,
0.03387451171875,
0.019775390625,
0.01690673828125,
0.0085906982421875,
0.01113128662109375,
0.038055419921875,
-0.0252227783203125,
0.046142578125,
0.025634765625,
-0.0025806427001953125,
-0.05364990234375,
0.055633544921875,
0.0038013458251953125,
0.0164947509765625,
0.005168914794921875,
0.0157012939453125,
-0.0267486572265625,
-0.030364990234375,
-0.0279693603515625,
0.033050537109375,
-0.0279541015625,
-0.0228729248046875,
-0.032989501953125,
-0.00833892822265625,
-0.047027587890625,
-0.03094482421875,
-0.01058197021484375,
-0.0238800048828125,
-0.039398193359375,
-0.0198974609375,
0.03289794921875,
0.0347900390625,
-0.022064208984375,
0.01593017578125,
-0.0262298583984375,
0.0242156982421875,
0.0146331787109375,
0.0120086669921875,
-0.0006737709045410156,
-0.05499267578125,
-0.002063751220703125,
-0.0078277587890625,
-0.04083251953125,
-0.0650634765625,
0.048675537109375,
0.0031833648681640625,
0.051055908203125,
0.04180908203125,
-0.0012140274047851562,
0.07476806640625,
-0.0065765380859375,
0.0802001953125,
0.0279998779296875,
-0.054931640625,
0.05206298828125,
-0.024017333984375,
0.0187225341796875,
0.046478271484375,
0.03802490234375,
-0.01495361328125,
-0.02777099609375,
-0.054473876953125,
-0.0738525390625,
0.0770263671875,
0.022216796875,
-0.0286865234375,
0.0045318603515625,
0.0174713134765625,
-0.0183563232421875,
0.003322601318359375,
-0.0616455078125,
-0.058624267578125,
-0.01081085205078125,
-0.01739501953125,
-0.0125274658203125,
-0.004772186279296875,
-0.01435089111328125,
-0.0462646484375,
0.044677734375,
0.00795745849609375,
0.036773681640625,
0.0239715576171875,
0.008819580078125,
0.00530242919921875,
-0.002838134765625,
0.049346923828125,
0.050750732421875,
-0.0313720703125,
0.0013456344604492188,
0.009429931640625,
-0.058197021484375,
0.023345947265625,
0.00444793701171875,
-0.016204833984375,
-0.010833740234375,
0.042572021484375,
0.045074462890625,
-0.0007085800170898438,
-0.0284423828125,
0.039703369140625,
0.00936126708984375,
-0.04071044921875,
-0.036651611328125,
0.01119232177734375,
-0.00516510009765625,
0.0266571044921875,
0.041046142578125,
0.01068878173828125,
0.00554656982421875,
-0.0400390625,
0.004016876220703125,
0.034515380859375,
-0.0178680419921875,
-0.004245758056640625,
0.0709228515625,
-0.0112152099609375,
-0.004322052001953125,
0.036712646484375,
-0.007785797119140625,
-0.0341796875,
0.07550048828125,
0.0287322998046875,
0.032989501953125,
-0.021087646484375,
0.00348663330078125,
0.0748291015625,
0.0308837890625,
-0.0033416748046875,
0.04071044921875,
0.005794525146484375,
-0.0269012451171875,
0.003772735595703125,
-0.051727294921875,
-0.01123809814453125,
0.0169219970703125,
-0.058258056640625,
0.0198516845703125,
-0.039337158203125,
-0.0246429443359375,
0.00945281982421875,
0.022552490234375,
-0.049163818359375,
0.030120849609375,
-0.0060882568359375,
0.07037353515625,
-0.06903076171875,
0.060821533203125,
0.04681396484375,
-0.06414794921875,
-0.0927734375,
-0.02423095703125,
0.0018186569213867188,
-0.05859375,
0.05908203125,
0.00870513916015625,
0.014312744140625,
-0.004825592041015625,
-0.0267333984375,
-0.09869384765625,
0.1165771484375,
0.0035457611083984375,
-0.02496337890625,
0.01367950439453125,
0.018280029296875,
0.0309906005859375,
0.007358551025390625,
0.038421630859375,
0.041259765625,
0.055450439453125,
0.006404876708984375,
-0.05877685546875,
0.02880859375,
-0.0328369140625,
-0.0126953125,
0.01412200927734375,
-0.07611083984375,
0.09112548828125,
-0.0244140625,
0.007091522216796875,
-0.00942230224609375,
0.044677734375,
0.046905517578125,
0.0200042724609375,
0.0230865478515625,
0.06146240234375,
0.0697021484375,
-0.021636962890625,
0.06756591796875,
-0.0178375244140625,
0.041412353515625,
0.0526123046875,
0.004856109619140625,
0.05322265625,
0.04388427734375,
-0.04443359375,
0.036468505859375,
0.0672607421875,
-0.00904083251953125,
0.042022705078125,
-0.0010957717895507812,
-0.00988006591796875,
-0.0005788803100585938,
0.0266571044921875,
-0.040252685546875,
0.00981903076171875,
0.0218658447265625,
-0.02716064453125,
-0.0015697479248046875,
-0.0192108154296875,
0.0207977294921875,
-0.0182037353515625,
-0.029388427734375,
0.029388427734375,
-0.0096435546875,
-0.056732177734375,
0.065673828125,
-0.007495880126953125,
0.04425048828125,
-0.03973388671875,
-0.0015516281127929688,
-0.0307464599609375,
0.033905029296875,
-0.03533935546875,
-0.07232666015625,
0.013458251953125,
0.00433349609375,
-0.015533447265625,
0.00627899169921875,
0.0159759521484375,
-0.0026226043701171875,
-0.0406494140625,
0.0187225341796875,
0.002391815185546875,
0.01436614990234375,
0.028411865234375,
-0.05364990234375,
0.01004791259765625,
0.0237579345703125,
-0.046661376953125,
0.019622802734375,
0.028778076171875,
-0.001575469970703125,
0.041015625,
0.060577392578125,
-0.0022716522216796875,
0.02423095703125,
-0.031707763671875,
0.07928466796875,
-0.060302734375,
-0.038787841796875,
-0.05487060546875,
0.038543701171875,
-0.0233306884765625,
-0.061676025390625,
0.06805419921875,
0.0718994140625,
0.03680419921875,
-0.0014438629150390625,
0.041015625,
-0.0469970703125,
0.028564453125,
-0.027496337890625,
0.048370361328125,
-0.057373046875,
0.0067901611328125,
-0.0240936279296875,
-0.042816162109375,
-0.0311279296875,
0.06475830078125,
-0.03759765625,
0.0032749176025390625,
0.0643310546875,
0.0687255859375,
0.009429931640625,
-0.006908416748046875,
-0.01045989990234375,
0.023529052734375,
0.0193328857421875,
0.0557861328125,
0.0251922607421875,
-0.04022216796875,
0.04327392578125,
-0.03643798828125,
-0.0280609130859375,
-0.0176239013671875,
-0.04583740234375,
-0.0667724609375,
-0.03875732421875,
-0.02703857421875,
-0.03448486328125,
-0.006725311279296875,
0.05462646484375,
0.0357666015625,
-0.0528564453125,
-0.026763916015625,
-0.00237274169921875,
0.003765106201171875,
-0.030181884765625,
-0.017242431640625,
0.07025146484375,
-0.00759124755859375,
-0.04583740234375,
-0.0055999755859375,
-0.0123748779296875,
0.0026397705078125,
0.005817413330078125,
-0.0261077880859375,
-0.034210205078125,
0.00901031494140625,
0.0248870849609375,
0.02423095703125,
-0.050750732421875,
-0.016265869140625,
-0.0047760009765625,
-0.025726318359375,
0.0274200439453125,
0.003856658935546875,
-0.0291748046875,
0.0006499290466308594,
0.0289154052734375,
0.01215362548828125,
0.06304931640625,
0.0019502639770507812,
-0.0140380859375,
-0.024993896484375,
0.02288818359375,
-0.00189208984375,
0.024383544921875,
-0.006221771240234375,
-0.0269622802734375,
0.03985595703125,
0.030120849609375,
-0.040496826171875,
-0.07049560546875,
-0.0279693603515625,
-0.099365234375,
-0.0191802978515625,
0.0843505859375,
-0.01052093505859375,
-0.039093017578125,
0.0086517333984375,
-0.0241851806640625,
0.0084686279296875,
-0.0435791015625,
0.036285400390625,
0.0465087890625,
-0.033050537109375,
-0.0006208419799804688,
-0.045379638671875,
0.0260467529296875,
0.0177459716796875,
-0.061859130859375,
-0.0179901123046875,
0.033538818359375,
0.0318603515625,
0.023162841796875,
0.0517578125,
-0.01331329345703125,
0.020751953125,
0.029022216796875,
0.00959014892578125,
-0.0046234130859375,
0.0078887939453125,
-0.0005211830139160156,
0.0091705322265625,
-0.00200653076171875,
-0.03839111328125
]
] |
facebook/deit-tiny-patch16-224 | 2022-07-13T11:53:31.000Z | [
"transformers",
"pytorch",
"tf",
"vit",
"image-classification",
"dataset:imagenet",
"arxiv:2012.12877",
"arxiv:2006.03677",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | image-classification | facebook | null | null | facebook/deit-tiny-patch16-224 | 1 | 7,432 | transformers | 2022-03-02T23:29:05 | ---
license: apache-2.0
tags:
- image-classification
datasets:
- imagenet
---
# Data-efficient Image Transformer (tiny-sized model)
Data-efficient Image Transformer (DeiT) model pre-trained and fine-tuned on ImageNet-1k (1 million images, 1,000 classes) at resolution 224x224. It was first introduced in the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Touvron et al. and first released in [this repository](https://github.com/facebookresearch/deit). However, the weights were converted from the [timm repository](https://github.com/rwightman/pytorch-image-models) by Ross Wightman.
Disclaimer: The team releasing DeiT did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
This model is actually a more efficiently trained Vision Transformer (ViT).
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pre-trained and fine-tuned on a large collection of images in a supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=facebook/deit) to look for
fine-tuned versions on a task that interests you.
### How to use
Since this model is a more efficiently trained ViT model, you can plug it into ViTModel or ViTForImageClassification. Note that the model expects the data to be prepared using DeiTFeatureExtractor. Here we use AutoFeatureExtractor, which will automatically use the appropriate feature extractor given the model name.
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoFeatureExtractor, ViTForImageClassification
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
feature_extractor = AutoFeatureExtractor.from_pretrained('facebook/deit-tiny-patch16-224')
model = ViTForImageClassification.from_pretrained('facebook/deit-tiny-patch16-224')
inputs = feature_extractor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
Currently, both the feature extractor and model support PyTorch. Tensorflow and JAX/FLAX are coming soon.
## Training data
The ViT model was pretrained on [ImageNet-1k](http://www.image-net.org/challenges/LSVRC/2012/), a dataset consisting of 1 million images and 1k classes.
## Training procedure
### Preprocessing
The exact details of preprocessing of images during training/validation can be found [here](https://github.com/facebookresearch/deit/blob/ab5715372db8c6cad5740714b2216d55aeae052e/datasets.py#L78).
At inference time, images are resized/rescaled to the same resolution (256x256), center-cropped at 224x224 and normalized across the RGB channels with the ImageNet mean and standard deviation.
### Pretraining
The model was trained on a single 8-GPU node for 3 days. Training resolution is 224. For all hyperparameters (such as batch size and learning rate) we refer to table 9 of the original paper.
## Evaluation results
| Model | ImageNet top-1 accuracy | ImageNet top-5 accuracy | # params | URL |
|---------------------------------------|-------------------------|-------------------------|----------|------------------------------------------------------------------|
| **DeiT-tiny** | **72.2** | **91.1** | **5M** | **https://huggingface.co/facebook/deit-tiny-patch16-224** |
| DeiT-small | 79.9 | 95.0 | 22M | https://huggingface.co/facebook/deit-small-patch16-224 |
| DeiT-base | 81.8 | 95.6 | 86M | https://huggingface.co/facebook/deit-base-patch16-224 |
| DeiT-tiny distilled | 74.5 | 91.9 | 6M | https://huggingface.co/facebook/deit-tiny-distilled-patch16-224 |
| DeiT-small distilled | 81.2 | 95.4 | 22M | https://huggingface.co/facebook/deit-small-distilled-patch16-224 |
| DeiT-base distilled | 83.4 | 96.5 | 87M | https://huggingface.co/facebook/deit-base-distilled-patch16-224 |
| DeiT-base 384 | 82.9 | 96.2 | 87M | https://huggingface.co/facebook/deit-base-patch16-384 |
| DeiT-base distilled 384 (1000 epochs) | 85.2 | 97.2 | 88M | https://huggingface.co/facebook/deit-base-distilled-patch16-384 |
Note that for fine-tuning, the best results are obtained with a higher resolution (384x384). Of course, increasing the model size will result in better performance.
### BibTeX entry and citation info
```bibtex
@misc{touvron2021training,
title={Training data-efficient image transformers & distillation through attention},
author={Hugo Touvron and Matthieu Cord and Matthijs Douze and Francisco Massa and Alexandre Sablayrolles and Hervé Jégou},
year={2021},
eprint={2012.12877},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@misc{wu2020visual,
title={Visual Transformers: Token-based Image Representation and Processing for Computer Vision},
author={Bichen Wu and Chenfeng Xu and Xiaoliang Dai and Alvin Wan and Peizhao Zhang and Zhicheng Yan and Masayoshi Tomizuka and Joseph Gonzalez and Kurt Keutzer and Peter Vajda},
year={2020},
eprint={2006.03677},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
```
```bibtex
@inproceedings{deng2009imagenet,
title={Imagenet: A large-scale hierarchical image database},
author={Deng, Jia and Dong, Wei and Socher, Richard and Li, Li-Jia and Li, Kai and Fei-Fei, Li},
booktitle={2009 IEEE conference on computer vision and pattern recognition},
pages={248--255},
year={2009},
organization={Ieee}
}
``` | 7,279 | [
[
-0.057037353515625,
-0.032745361328125,
0.004810333251953125,
0.0022869110107421875,
-0.02740478515625,
-0.0185546875,
-0.0089874267578125,
-0.038299560546875,
0.026031494140625,
0.0166778564453125,
-0.0286407470703125,
-0.0238800048828125,
-0.0606689453125,
-0.0002605915069580078,
-0.03302001953125,
0.06768798828125,
0.0034542083740234375,
-0.01100921630859375,
-0.007671356201171875,
-0.01174163818359375,
-0.032470703125,
-0.0343017578125,
-0.053375244140625,
-0.0105438232421875,
0.03839111328125,
0.017364501953125,
0.0509033203125,
0.0626220703125,
0.0650634765625,
0.03515625,
-0.007442474365234375,
0.00569915771484375,
-0.036224365234375,
-0.0252532958984375,
0.006183624267578125,
-0.0241241455078125,
-0.03106689453125,
0.0169830322265625,
0.039886474609375,
0.03228759765625,
0.0166778564453125,
0.0221405029296875,
0.019195556640625,
0.0543212890625,
-0.037017822265625,
0.0125579833984375,
-0.034515380859375,
0.01751708984375,
-0.00308990478515625,
-0.00457763671875,
-0.020538330078125,
-0.01751708984375,
0.01654052734375,
-0.037384033203125,
0.0279998779296875,
-0.01030731201171875,
0.1044921875,
0.03057861328125,
-0.02374267578125,
0.013702392578125,
-0.046173095703125,
0.05389404296875,
-0.0350341796875,
0.029449462890625,
0.02386474609375,
0.0253753662109375,
0.006755828857421875,
-0.07647705078125,
-0.038238525390625,
-0.008148193359375,
-0.0247344970703125,
0.013671875,
-0.0248870849609375,
0.0031719207763671875,
0.039154052734375,
0.04669189453125,
-0.034423828125,
0.0013837814331054688,
-0.036834716796875,
-0.0165557861328125,
0.0484619140625,
-0.01038360595703125,
0.006591796875,
-0.008087158203125,
-0.0477294921875,
-0.024383544921875,
-0.0166168212890625,
0.008880615234375,
0.004566192626953125,
0.004299163818359375,
-0.017669677734375,
0.03509521484375,
-0.008087158203125,
0.044647216796875,
0.035400390625,
-0.0012578964233398438,
0.04217529296875,
-0.0241546630859375,
-0.0295867919921875,
-0.0089111328125,
0.071044921875,
0.03399658203125,
0.021759033203125,
0.009979248046875,
-0.0180206298828125,
0.00716400146484375,
0.01995849609375,
-0.08251953125,
-0.0222320556640625,
-0.0013990402221679688,
-0.0572509765625,
-0.036285400390625,
0.0189666748046875,
-0.046630859375,
-0.0037899017333984375,
-0.0276641845703125,
0.04248046875,
-0.02984619140625,
-0.0240325927734375,
-0.012176513671875,
-0.0034465789794921875,
0.029937744140625,
0.0244598388671875,
-0.04254150390625,
0.014862060546875,
0.0198211669921875,
0.076171875,
-0.006809234619140625,
-0.0117645263671875,
-0.006275177001953125,
-0.0277862548828125,
-0.0340576171875,
0.045867919921875,
-0.0017900466918945312,
-0.0013942718505859375,
-0.0152587890625,
0.0271759033203125,
-0.006488800048828125,
-0.03271484375,
0.024566650390625,
-0.033966064453125,
-0.0027980804443359375,
-0.0194854736328125,
-0.0240020751953125,
-0.01708984375,
0.0234832763671875,
-0.061676025390625,
0.08831787109375,
0.0203857421875,
-0.06884765625,
0.0307769775390625,
-0.0389404296875,
-0.00582122802734375,
-0.00653076171875,
0.0011892318725585938,
-0.045440673828125,
-0.006061553955078125,
0.0244598388671875,
0.04705810546875,
-0.01419830322265625,
-0.004177093505859375,
-0.0196075439453125,
-0.036712646484375,
0.0197296142578125,
-0.03125,
0.0693359375,
0.0247344970703125,
-0.037445068359375,
-0.005527496337890625,
-0.058685302734375,
0.00505828857421875,
0.026824951171875,
-0.017822265625,
-0.00644683837890625,
-0.03424072265625,
0.0063629150390625,
0.03253173828125,
0.0179901123046875,
-0.04412841796875,
0.00795745849609375,
-0.01068115234375,
0.039703369140625,
0.063232421875,
-0.01003265380859375,
0.0287017822265625,
-0.0227203369140625,
0.0220184326171875,
0.022674560546875,
0.0330810546875,
-0.021514892578125,
-0.033355712890625,
-0.0714111328125,
-0.029449462890625,
0.035064697265625,
0.029693603515625,
-0.0523681640625,
0.0479736328125,
-0.03302001953125,
-0.04888916015625,
-0.02789306640625,
0.00357818603515625,
0.02685546875,
0.03631591796875,
0.03204345703125,
-0.038726806640625,
-0.038055419921875,
-0.0797119140625,
0.0079193115234375,
-0.003704071044921875,
0.01172637939453125,
0.014495849609375,
0.052703857421875,
-0.018310546875,
0.0732421875,
-0.032318115234375,
-0.025421142578125,
-0.00162506103515625,
-0.00658416748046875,
0.0247650146484375,
0.052520751953125,
0.0643310546875,
-0.07293701171875,
-0.0518798828125,
0.0029163360595703125,
-0.062347412109375,
0.018341064453125,
0.0008020401000976562,
-0.028472900390625,
0.006908416748046875,
0.031097412109375,
-0.04876708984375,
0.062286376953125,
0.0223541259765625,
-0.01284027099609375,
0.02508544921875,
-0.006496429443359375,
0.0187530517578125,
-0.081787109375,
0.004791259765625,
0.026641845703125,
-0.0287322998046875,
-0.036895751953125,
-0.003814697265625,
0.007320404052734375,
-0.000457763671875,
-0.04010009765625,
0.0248565673828125,
-0.042205810546875,
-0.00907135009765625,
-0.0107879638671875,
-0.02264404296875,
0.001102447509765625,
0.048614501953125,
0.0014123916625976562,
0.0428466796875,
0.047637939453125,
-0.037811279296875,
0.041534423828125,
0.02166748046875,
-0.0288543701171875,
0.047943115234375,
-0.06085205078125,
0.0186767578125,
-0.00933837890625,
0.0223846435546875,
-0.0814208984375,
-0.018402099609375,
0.01488494873046875,
-0.041229248046875,
0.040863037109375,
-0.019775390625,
-0.028045654296875,
-0.061309814453125,
-0.0240020751953125,
0.033111572265625,
0.05169677734375,
-0.052825927734375,
0.0278167724609375,
0.012420654296875,
0.0294647216796875,
-0.0537109375,
-0.07879638671875,
-0.0068359375,
-0.019805908203125,
-0.046356201171875,
0.038330078125,
0.003566741943359375,
0.01190948486328125,
0.015167236328125,
0.0008206367492675781,
-0.01544952392578125,
-0.0126800537109375,
0.034027099609375,
0.0274658203125,
-0.02105712890625,
-0.0028057098388671875,
-0.02618408203125,
-0.01690673828125,
-0.007183074951171875,
-0.0338134765625,
0.0322265625,
-0.031463623046875,
-0.0209197998046875,
-0.062286376953125,
0.00438690185546875,
0.049560546875,
-0.0085906982421875,
0.049774169921875,
0.06494140625,
-0.03955078125,
0.008392333984375,
-0.045684814453125,
-0.0189971923828125,
-0.03924560546875,
0.0252532958984375,
-0.0303192138671875,
-0.044830322265625,
0.052581787109375,
0.007503509521484375,
0.0013561248779296875,
0.057373046875,
0.0330810546875,
-0.017974853515625,
0.07025146484375,
0.042236328125,
-0.005096435546875,
0.05859375,
-0.065185546875,
0.003326416015625,
-0.05023193359375,
-0.0134735107421875,
-0.016937255859375,
-0.053558349609375,
-0.050201416015625,
-0.0296783447265625,
0.021759033203125,
0.0040435791015625,
-0.033355712890625,
0.048309326171875,
-0.06207275390625,
0.0247039794921875,
0.058441162109375,
0.040771484375,
-0.00827789306640625,
0.021759033203125,
-0.006748199462890625,
-0.0037097930908203125,
-0.04632568359375,
-0.0108642578125,
0.06781005859375,
0.034759521484375,
0.05126953125,
-0.0198974609375,
0.045318603515625,
0.01081085205078125,
0.011566162109375,
-0.05694580078125,
0.042236328125,
-0.0157928466796875,
-0.04962158203125,
-0.007175445556640625,
-0.0271759033203125,
-0.07098388671875,
0.00968170166015625,
-0.0156402587890625,
-0.04705810546875,
0.041015625,
0.024322509765625,
-0.0203857421875,
0.0377197265625,
-0.05548095703125,
0.065673828125,
-0.0176544189453125,
-0.03155517578125,
0.00945281982421875,
-0.060699462890625,
0.01019287109375,
0.00018477439880371094,
-0.0064849853515625,
0.012969970703125,
0.0201568603515625,
0.057037353515625,
-0.0594482421875,
0.07025146484375,
-0.0229339599609375,
0.0211944580078125,
0.05059814453125,
-0.0177459716796875,
0.0265960693359375,
-0.0198974609375,
0.0091552734375,
0.033477783203125,
0.0001533031463623047,
-0.03656005859375,
-0.036224365234375,
0.044677734375,
-0.0692138671875,
-0.0232086181640625,
-0.035308837890625,
-0.0199432373046875,
0.01259613037109375,
0.015380859375,
0.05517578125,
0.040069580078125,
0.0063934326171875,
0.0413818359375,
0.048095703125,
-0.019989013671875,
0.0360107421875,
-0.0177001953125,
0.00093841552734375,
-0.02764892578125,
0.067138671875,
0.03094482421875,
0.0184326171875,
0.022674560546875,
0.016265869140625,
-0.01898193359375,
-0.0284881591796875,
-0.02569580078125,
0.01157379150390625,
-0.061065673828125,
-0.03900146484375,
-0.048858642578125,
-0.04730224609375,
-0.031463623046875,
-0.01096343994140625,
-0.04608154296875,
-0.0283966064453125,
-0.0328369140625,
-0.01039886474609375,
0.044403076171875,
0.045867919921875,
-0.0219879150390625,
0.03839111328125,
-0.0452880859375,
0.01446533203125,
0.03082275390625,
0.0294647216796875,
-0.004047393798828125,
-0.052734375,
-0.026336669921875,
0.01058197021484375,
-0.0244293212890625,
-0.052001953125,
0.02899169921875,
0.0207977294921875,
0.04119873046875,
0.039794921875,
-0.011871337890625,
0.072998046875,
-0.0195770263671875,
0.0523681640625,
0.034942626953125,
-0.046478271484375,
0.05108642578125,
-0.01248931884765625,
0.01055908203125,
0.03765869140625,
0.0352783203125,
-0.0191650390625,
0.0012826919555664062,
-0.05743408203125,
-0.057586669921875,
0.053009033203125,
0.01317596435546875,
0.0083770751953125,
0.00785064697265625,
0.044830322265625,
-0.01503753662109375,
0.0015811920166015625,
-0.06182861328125,
-0.031768798828125,
-0.037445068359375,
-0.0153961181640625,
0.0025959014892578125,
-0.01534271240234375,
0.007404327392578125,
-0.057037353515625,
0.0419921875,
-0.00701904296875,
0.055267333984375,
0.0185546875,
-0.0158233642578125,
0.0015506744384765625,
-0.032379150390625,
0.01284027099609375,
0.031524658203125,
-0.014923095703125,
0.00444793701171875,
0.00795745849609375,
-0.05712890625,
0.00777435302734375,
0.0056304931640625,
-0.01140594482421875,
-0.0019273757934570312,
0.032012939453125,
0.0755615234375,
-0.0008611679077148438,
-0.0010862350463867188,
0.0582275390625,
-0.0116424560546875,
-0.034454345703125,
-0.0306854248046875,
0.002696990966796875,
-0.01250457763671875,
0.028656005859375,
0.0278167724609375,
0.0224456787109375,
0.006793975830078125,
-0.023406982421875,
0.02838134765625,
0.02557373046875,
-0.0419921875,
-0.0244903564453125,
0.049224853515625,
-0.007213592529296875,
0.0078582763671875,
0.057403564453125,
-0.006927490234375,
-0.043853759765625,
0.07861328125,
0.0287933349609375,
0.06134033203125,
-0.020111083984375,
0.01143646240234375,
0.06024169921875,
0.0201416015625,
-0.01050567626953125,
0.005077362060546875,
0.0041351318359375,
-0.053314208984375,
-0.018310546875,
-0.049713134765625,
0.018829345703125,
0.0211639404296875,
-0.054046630859375,
0.0265655517578125,
-0.036041259765625,
-0.04058837890625,
0.0190887451171875,
0.0028247833251953125,
-0.083740234375,
0.0296630859375,
0.0121002197265625,
0.0634765625,
-0.05999755859375,
0.0582275390625,
0.057220458984375,
-0.044677734375,
-0.07550048828125,
-0.0215301513671875,
-0.003711700439453125,
-0.05584716796875,
0.06402587890625,
0.031494140625,
0.01203155517578125,
0.01338958740234375,
-0.056488037109375,
-0.07196044921875,
0.099853515625,
0.02252197265625,
-0.032318115234375,
0.0010709762573242188,
0.0103912353515625,
0.033599853515625,
-0.0232391357421875,
0.040283203125,
0.0252685546875,
0.0272979736328125,
0.03204345703125,
-0.0594482421875,
0.0061798095703125,
-0.032745361328125,
0.02593994140625,
-0.0033111572265625,
-0.0660400390625,
0.077880859375,
-0.00925445556640625,
-0.00998687744140625,
-0.0036907196044921875,
0.05059814453125,
-0.01043701171875,
-0.0022125244140625,
0.05755615234375,
0.06036376953125,
0.0294647216796875,
-0.0233001708984375,
0.07733154296875,
-0.005161285400390625,
0.0423583984375,
0.04754638671875,
0.027801513671875,
0.032257080078125,
0.0277099609375,
-0.023345947265625,
0.02667236328125,
0.08319091796875,
-0.025299072265625,
0.043060302734375,
0.00122833251953125,
0.00785064697265625,
-0.0126800537109375,
0.000003874301910400391,
-0.037567138671875,
0.03857421875,
0.0168609619140625,
-0.05078125,
-0.0038471221923828125,
0.019683837890625,
-0.01314544677734375,
-0.027679443359375,
-0.028045654296875,
0.051849365234375,
0.00586700439453125,
-0.034027099609375,
0.06304931640625,
-0.0158538818359375,
0.056640625,
-0.0222625732421875,
-0.0036525726318359375,
-0.01995849609375,
0.0310821533203125,
-0.0257415771484375,
-0.055419921875,
0.0174560546875,
-0.006591796875,
-0.0033779144287109375,
-0.0092315673828125,
0.0699462890625,
-0.01305389404296875,
-0.045654296875,
0.0213470458984375,
0.018310546875,
0.0219879150390625,
-0.007049560546875,
-0.0712890625,
0.0013561248779296875,
-0.0020618438720703125,
-0.050750732421875,
0.0189971923828125,
0.035308837890625,
-0.0011606216430664062,
0.0341796875,
0.05010986328125,
-0.005157470703125,
0.02099609375,
-0.01374053955078125,
0.08477783203125,
-0.034942626953125,
-0.0361328125,
-0.05694580078125,
0.046356201171875,
-0.0219879150390625,
-0.0225982666015625,
0.044036865234375,
0.03271484375,
0.0721435546875,
-0.004581451416015625,
0.0501708984375,
-0.02557373046875,
0.00951385498046875,
-0.017486572265625,
0.04010009765625,
-0.055267333984375,
-0.0160980224609375,
-0.0296173095703125,
-0.07757568359375,
-0.0182342529296875,
0.07403564453125,
-0.0117034912109375,
0.030914306640625,
0.03973388671875,
0.05792236328125,
-0.0247344970703125,
-0.0149383544921875,
0.0173797607421875,
0.0164337158203125,
0.01351165771484375,
0.035064697265625,
0.051849365234375,
-0.059234619140625,
0.038055419921875,
-0.0604248046875,
-0.0271453857421875,
-0.02191162109375,
-0.05694580078125,
-0.07281494140625,
-0.0616455078125,
-0.039215087890625,
-0.03619384765625,
-0.0097503662109375,
0.060821533203125,
0.0814208984375,
-0.051727294921875,
0.00498199462890625,
-0.00978851318359375,
-0.019805908203125,
-0.029266357421875,
-0.01715087890625,
0.0435791015625,
-0.0020198822021484375,
-0.0643310546875,
-0.0206451416015625,
-0.0014162063598632812,
0.026641845703125,
-0.012939453125,
-0.00455474853515625,
-0.0190582275390625,
-0.02154541015625,
0.0460205078125,
0.015716552734375,
-0.0311431884765625,
-0.032012939453125,
0.0060882568359375,
-0.01739501953125,
0.0236663818359375,
0.04052734375,
-0.048858642578125,
0.0247955322265625,
0.043548583984375,
0.0440673828125,
0.06207275390625,
0.004852294921875,
0.0006918907165527344,
-0.053497314453125,
0.03326416015625,
0.0032405853271484375,
0.0340576171875,
0.020751953125,
-0.040069580078125,
0.053955078125,
0.035400390625,
-0.044342041015625,
-0.05511474609375,
-0.006427764892578125,
-0.09100341796875,
-0.0105133056640625,
0.07562255859375,
-0.0270233154296875,
-0.040191650390625,
0.0236663818359375,
-0.01096343994140625,
0.039093017578125,
-0.0114898681640625,
0.032073974609375,
0.03564453125,
0.0022220611572265625,
-0.029541015625,
-0.050384521484375,
0.0264129638671875,
0.0037212371826171875,
-0.044830322265625,
-0.0274200439453125,
0.0241241455078125,
0.02423095703125,
0.032073974609375,
0.044403076171875,
-0.0247039794921875,
0.007526397705078125,
0.00623321533203125,
0.01409912109375,
-0.01222991943359375,
-0.0225067138671875,
-0.01284027099609375,
-0.0098876953125,
-0.0216064453125,
-0.050506591796875
]
] |
LiYuan/amazon-review-sentiment-analysis | 2022-04-30T22:03:23.000Z | [
"transformers",
"pytorch",
"tensorboard",
"bert",
"text-classification",
"generated_from_trainer",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | text-classification | LiYuan | null | null | LiYuan/amazon-review-sentiment-analysis | 18 | 7,431 | transformers | 2022-04-30T20:37:44 | ---
license: apache-2.0
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: distilbert-base-uncased-finetuned-mnli-amazon-query-shopping
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. -->
# distilbert-base-uncased-finetuned-mnli-amazon-query-shopping
This model is a fine-tuned version of [nlptown/bert-base-multilingual-uncased-sentiment](https://huggingface.co/nlptown/bert-base-multilingual-uncased-sentiment?text=I+like+you.+I+love+you) on an [Amazon US Customer Reviews Dataset](https://www.kaggle.com/datasets/cynthiarempel/amazon-us-customer-reviews-dataset). The code for the fine-tuning process can be found
[here](https://github.com/vanderbilt-data-science/bigdata/blob/main/06-fine-tune-BERT-on-our-dataset.ipynb). This model is uncased: it does
not make a difference between english and English.
It achieves the following results on the evaluation set:
- Loss: 0.5202942490577698
- Accuracy: 0.8
## Model description
This a bert-base-multilingual-uncased model finetuned for sentiment analysis on product reviews in six languages: English, Dutch, German, French, Spanish and Italian. It predicts the sentiment of the review as a number of stars (between 1 and 5).
This model is intended for direct use as a sentiment analysis model for product reviews in any of the six languages above, or for further finetuning on related sentiment analysis tasks.
We replaced its head with our customer reviews to fine-tune it on 17,280 rows of training set while validating it on 4,320 rows of dev set. Finally, we evaluated our model performance on a held-out test set: 2,400 rows.
## Intended uses & limitations
Bert-base is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked)
to make decisions, such as sequence classification, token classification, or question answering. This fine-tuned version of BERT-base is used to predict review rating star given the review.
The limitations are this trained model is focusing on reviews and products on Amazon. If you apply this model to other domains, it may perform poorly.
## How to use
You can use this model directly by downloading the trained weights and configurations like the below code snippet:
```python
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("LiYuan/amazon-review-sentiment-analysis")
model = AutoModelForSequenceClassification.from_pretrained("LiYuan/amazon-review-sentiment-analysis")
```
## Training and evaluation data
Download all the raw [dataset](https://www.kaggle.com/datasets/cynthiarempel/amazon-us-customer-reviews-dataset) from the Kaggle website.
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 0.555400 | 1.0 | 1080 | 0.520294 | 0.800000 |
| 0.424300 | 2.0 | 1080 | 0.549649 | 0.798380 |
### Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.1.0
- Tokenizers 0.12.1 | 3,515 | [
[
-0.045745849609375,
-0.057281494140625,
-0.003170013427734375,
0.0277252197265625,
-0.0285491943359375,
-0.018951416015625,
-0.026214599609375,
-0.034698486328125,
0.011810302734375,
0.031646728515625,
-0.048797607421875,
-0.043701171875,
-0.041839599609375,
-0.0032787322998046875,
-0.006671905517578125,
0.1136474609375,
0.012542724609375,
0.031463623046875,
-0.000492095947265625,
-0.01326751708984375,
-0.01171112060546875,
-0.06317138671875,
-0.0364990234375,
-0.0254364013671875,
0.0290374755859375,
0.0286407470703125,
0.06463623046875,
0.01488494873046875,
0.03424072265625,
0.01468658447265625,
-0.0160980224609375,
0.0089263916015625,
-0.033294677734375,
-0.0193328857421875,
0.0118408203125,
-0.039154052734375,
-0.048431396484375,
0.005462646484375,
0.02734375,
0.048858642578125,
0.005222320556640625,
0.030059814453125,
0.022491455078125,
0.058990478515625,
-0.027252197265625,
0.0251617431640625,
-0.038604736328125,
0.016387939453125,
0.019195556640625,
-0.004901885986328125,
-0.019866943359375,
-0.0279693603515625,
0.016815185546875,
-0.031097412109375,
0.033477783203125,
-0.00353240966796875,
0.08880615234375,
0.00469970703125,
-0.0216064453125,
-0.0171966552734375,
-0.06756591796875,
0.0655517578125,
-0.0733642578125,
0.0209197998046875,
0.0247650146484375,
0.01123809814453125,
0.006542205810546875,
-0.0263214111328125,
-0.057342529296875,
-0.0175018310546875,
-0.0181884765625,
0.0204925537109375,
-0.0152435302734375,
0.002696990966796875,
0.0357666015625,
0.04681396484375,
-0.03961181640625,
0.005855560302734375,
-0.035736083984375,
-0.00850677490234375,
0.058349609375,
0.0008487701416015625,
-0.0128936767578125,
-0.044342041015625,
-0.041412353515625,
-0.0184478759765625,
-0.03094482421875,
0.037200927734375,
0.051055908203125,
0.0218505859375,
-0.00879669189453125,
0.03466796875,
-0.0305938720703125,
0.03350830078125,
0.0237579345703125,
0.004947662353515625,
0.05413818359375,
-0.0074615478515625,
-0.0282745361328125,
-0.0164947509765625,
0.079345703125,
0.052154541015625,
0.0261383056640625,
0.01506805419921875,
-0.0294952392578125,
0.0017213821411132812,
0.0163116455078125,
-0.053802490234375,
-0.01465606689453125,
0.023406982421875,
-0.049407958984375,
-0.041046142578125,
0.0037364959716796875,
-0.045196533203125,
0.004787445068359375,
-0.033294677734375,
0.044464111328125,
-0.027313232421875,
-0.01214599609375,
0.01457977294921875,
0.0077667236328125,
0.0119781494140625,
-0.0003654956817626953,
-0.06341552734375,
0.01050567626953125,
0.03277587890625,
0.0223541259765625,
0.0015707015991210938,
-0.03387451171875,
0.00885009765625,
-0.027862548828125,
-0.024566650390625,
0.0447998046875,
-0.005115509033203125,
-0.019775390625,
0.0154266357421875,
0.029632568359375,
0.00969696044921875,
-0.034820556640625,
0.056671142578125,
-0.042449951171875,
0.032379150390625,
-0.0227508544921875,
-0.047393798828125,
-0.0364990234375,
0.038299560546875,
-0.03765869140625,
0.08807373046875,
0.0016613006591796875,
-0.034637451171875,
0.0411376953125,
-0.03106689453125,
-0.029693603515625,
-0.0227813720703125,
0.0159149169921875,
-0.0577392578125,
0.002452850341796875,
0.03125,
0.04736328125,
0.001953125,
0.017303466796875,
-0.024261474609375,
-0.0272674560546875,
0.0151824951171875,
-0.03369140625,
0.07037353515625,
0.0178375244140625,
-0.038238525390625,
-0.0022335052490234375,
-0.06842041015625,
0.01140594482421875,
0.0206756591796875,
-0.03350830078125,
-0.000301361083984375,
-0.0157928466796875,
0.030670166015625,
0.0174102783203125,
0.032318115234375,
-0.044036865234375,
0.0236663818359375,
-0.042144775390625,
0.0225677490234375,
0.038330078125,
0.00608062744140625,
0.022613525390625,
-0.0238800048828125,
0.033203125,
0.0203704833984375,
0.045196533203125,
0.004913330078125,
-0.0299072265625,
-0.08575439453125,
0.01873779296875,
0.0308837890625,
0.031463623046875,
-0.0295257568359375,
0.07427978515625,
-0.0121002197265625,
-0.044586181640625,
-0.03875732421875,
0.0197601318359375,
0.03009033203125,
0.0293426513671875,
0.033050537109375,
-0.0247039794921875,
-0.03656005859375,
-0.078857421875,
0.0005931854248046875,
-0.01390838623046875,
0.0033817291259765625,
0.0240325927734375,
0.0313720703125,
-0.024749755859375,
0.05426025390625,
-0.033233642578125,
-0.036529541015625,
-0.025177001953125,
0.016357421875,
0.03955078125,
0.04205322265625,
0.052276611328125,
-0.037933349609375,
-0.043121337890625,
-0.0174102783203125,
-0.04998779296875,
0.01107025146484375,
-0.01494598388671875,
-0.01161956787109375,
0.0303192138671875,
0.02069091796875,
-0.045745849609375,
0.0242919921875,
0.039306640625,
-0.0106201171875,
0.0501708984375,
-0.02105712890625,
0.002506256103515625,
-0.088134765625,
-0.0013551712036132812,
0.0093536376953125,
0.005840301513671875,
-0.023162841796875,
-0.01145172119140625,
0.01334381103515625,
-0.01371002197265625,
-0.021697998046875,
0.0171966552734375,
-0.0189666748046875,
0.006778717041015625,
-0.0134429931640625,
-0.0201416015625,
0.02398681640625,
0.08026123046875,
0.011749267578125,
0.035186767578125,
0.028961181640625,
-0.042999267578125,
0.022979736328125,
0.03118896484375,
-0.0433349609375,
0.036590576171875,
-0.0653076171875,
-0.0006814002990722656,
-0.00762176513671875,
0.016082763671875,
-0.07733154296875,
0.00027441978454589844,
0.0243377685546875,
-0.051422119140625,
0.0175628662109375,
-0.011962890625,
-0.054840087890625,
-0.03631591796875,
-0.0156097412109375,
0.00496673583984375,
0.0491943359375,
-0.04736328125,
0.023712158203125,
0.01114654541015625,
0.001125335693359375,
-0.058807373046875,
-0.057037353515625,
-0.024658203125,
-0.004589080810546875,
-0.03765869140625,
0.01386260986328125,
-0.015380859375,
0.0146331787109375,
-0.006107330322265625,
0.004985809326171875,
0.0001456737518310547,
-0.02001953125,
0.005970001220703125,
0.0230255126953125,
-0.0114898681640625,
0.0267181396484375,
0.004238128662109375,
-0.0100555419921875,
0.023040771484375,
-0.0037250518798828125,
0.04522705078125,
-0.0247039794921875,
-0.0091705322265625,
-0.0389404296875,
0.00970458984375,
0.0433349609375,
-0.01047515869140625,
0.04278564453125,
0.0654296875,
-0.020263671875,
-0.02392578125,
-0.0543212890625,
-0.0281829833984375,
-0.033203125,
0.03985595703125,
-0.01244354248046875,
-0.026702880859375,
0.043914794921875,
0.0283660888671875,
0.01352691650390625,
0.054595947265625,
0.04248046875,
-0.0243072509765625,
0.09197998046875,
0.041656494140625,
-0.02392578125,
0.03192138671875,
-0.049163818359375,
0.0234375,
-0.05181884765625,
-0.01800537109375,
-0.02484130859375,
-0.038604736328125,
-0.0494384765625,
0.005062103271484375,
0.015960693359375,
0.0240325927734375,
-0.031829833984375,
0.028411865234375,
-0.059539794921875,
0.0165557861328125,
0.055633544921875,
0.0218048095703125,
0.01374053955078125,
0.02325439453125,
-0.01629638671875,
-0.01099395751953125,
-0.045562744140625,
-0.03070068359375,
0.099853515625,
0.04083251953125,
0.05889892578125,
-0.0201416015625,
0.03070068359375,
0.0205230712890625,
0.0007085800170898438,
-0.0653076171875,
0.031219482421875,
-0.01313018798828125,
-0.0640869140625,
0.0001850128173828125,
-0.01416778564453125,
-0.06329345703125,
0.004268646240234375,
-0.0308837890625,
-0.0229339599609375,
0.0283355712890625,
0.0097503662109375,
-0.03448486328125,
0.02484130859375,
-0.05615234375,
0.072265625,
-0.043670654296875,
-0.016143798828125,
-0.01482391357421875,
-0.042236328125,
-0.0028591156005859375,
0.0174713134765625,
-0.00597381591796875,
-0.012664794921875,
0.026031494140625,
0.0631103515625,
-0.033050537109375,
0.07012939453125,
-0.02056884765625,
0.007110595703125,
0.022674560546875,
-0.0095062255859375,
0.0303955078125,
0.006031036376953125,
0.0025177001953125,
0.037750244140625,
-0.0093994140625,
-0.029144287109375,
-0.0345458984375,
0.057525634765625,
-0.08880615234375,
-0.020843505859375,
-0.038421630859375,
-0.035003662109375,
-0.033355712890625,
0.0133514404296875,
0.0426025390625,
0.029693603515625,
-0.0044403076171875,
0.0297698974609375,
0.044708251953125,
-0.01529693603515625,
0.02459716796875,
0.041168212890625,
-0.00983428955078125,
-0.043121337890625,
0.06158447265625,
0.0098114013671875,
0.006954193115234375,
0.01212310791015625,
0.01543426513671875,
-0.0343017578125,
-0.0323486328125,
-0.036865234375,
0.0210723876953125,
-0.0618896484375,
-0.0133819580078125,
-0.057891845703125,
-0.0313720703125,
-0.0256805419921875,
0.0031909942626953125,
-0.0264129638671875,
-0.0262298583984375,
-0.025970458984375,
-0.024078369140625,
0.0360107421875,
0.046630859375,
0.005374908447265625,
0.03021240234375,
-0.04925537109375,
0.00618743896484375,
0.0177154541015625,
0.039306640625,
0.006359100341796875,
-0.049774169921875,
-0.032470703125,
0.01474761962890625,
-0.041412353515625,
-0.0443115234375,
0.04052734375,
0.0009813308715820312,
0.035919189453125,
0.042938232421875,
0.0033702850341796875,
0.04986572265625,
-0.01708984375,
0.0689697265625,
0.0141143798828125,
-0.044586181640625,
0.038665771484375,
-0.03192138671875,
0.0214691162109375,
0.045440673828125,
0.054229736328125,
-0.0241241455078125,
-0.01386260986328125,
-0.06622314453125,
-0.0638427734375,
0.04815673828125,
0.0081024169921875,
0.033599853515625,
0.0015592575073242188,
0.0162200927734375,
0.01445770263671875,
0.038055419921875,
-0.08941650390625,
-0.038848876953125,
-0.01904296875,
-0.006664276123046875,
-0.02398681640625,
-0.03741455078125,
0.002605438232421875,
-0.05194091796875,
0.07403564453125,
0.00441741943359375,
0.0272064208984375,
0.01407623291015625,
-0.007793426513671875,
-0.01515960693359375,
0.016204833984375,
0.0131072998046875,
0.032684326171875,
-0.03765869140625,
-0.033203125,
0.0021724700927734375,
-0.0296783447265625,
-0.01629638671875,
0.0263519287109375,
-0.0251617431640625,
0.018157958984375,
-0.0026397705078125,
0.06829833984375,
0.01265716552734375,
-0.019805908203125,
0.03936767578125,
0.00083160400390625,
-0.025299072265625,
-0.04742431640625,
-0.0220489501953125,
-0.005832672119140625,
0.01326751708984375,
0.0243988037109375,
0.019134521484375,
0.01023101806640625,
-0.03863525390625,
0.00689697265625,
0.030670166015625,
-0.041534423828125,
-0.0211334228515625,
0.04339599609375,
0.0250396728515625,
-0.003238677978515625,
0.052520751953125,
-0.013427734375,
-0.042694091796875,
0.051025390625,
0.0280303955078125,
0.06884765625,
0.00286102294921875,
0.01519775390625,
0.058837890625,
0.0267181396484375,
0.00421142578125,
0.033050537109375,
0.0019683837890625,
-0.058319091796875,
-0.008331298828125,
-0.06695556640625,
-0.0279693603515625,
0.03863525390625,
-0.06732177734375,
0.0251617431640625,
-0.043060302734375,
-0.0290679931640625,
0.00839996337890625,
0.0220184326171875,
-0.061004638671875,
0.042236328125,
0.01483917236328125,
0.048980712890625,
-0.0660400390625,
0.053070068359375,
0.04541015625,
-0.054473876953125,
-0.06890869140625,
-0.006153106689453125,
-0.0193939208984375,
-0.0570068359375,
0.042327880859375,
0.026580810546875,
0.016937255859375,
-0.01220703125,
-0.0355224609375,
-0.0447998046875,
0.05426025390625,
-0.00606536865234375,
-0.04156494140625,
0.006855010986328125,
0.0265045166015625,
0.0655517578125,
-0.0272979736328125,
0.030975341796875,
0.0265960693359375,
0.00905609130859375,
-0.01482391357421875,
-0.060546875,
-0.0160064697265625,
-0.03887939453125,
-0.00569915771484375,
-0.00438690185546875,
-0.05517578125,
0.080322265625,
0.006351470947265625,
0.0253143310546875,
0.002765655517578125,
0.03350830078125,
0.006732940673828125,
0.011077880859375,
0.0386962890625,
0.055908203125,
0.033477783203125,
-0.0133056640625,
0.07196044921875,
-0.04803466796875,
0.0614013671875,
0.06512451171875,
0.0025463104248046875,
0.07476806640625,
0.0165863037109375,
-0.0207672119140625,
0.056182861328125,
0.05682373046875,
-0.038787841796875,
0.032806396484375,
-0.005970001220703125,
0.0011072158813476562,
-0.0140380859375,
0.00836944580078125,
-0.03973388671875,
0.0224456787109375,
0.01544952392578125,
-0.057037353515625,
-0.0020465850830078125,
0.0014190673828125,
0.0025463104248046875,
-0.0163421630859375,
-0.024261474609375,
0.051239013671875,
-0.00640106201171875,
-0.057342529296875,
0.072021484375,
-0.00827789306640625,
0.06591796875,
-0.051361083984375,
0.01114654541015625,
-0.017791748046875,
0.034637451171875,
-0.0289764404296875,
-0.057098388671875,
0.01087188720703125,
-0.00466156005859375,
-0.0289459228515625,
-0.0166015625,
0.032196044921875,
-0.02825927734375,
-0.06976318359375,
0.0245208740234375,
0.0308837890625,
0.0035991668701171875,
-0.00539398193359375,
-0.08197021484375,
-0.0143585205078125,
0.014617919921875,
-0.04248046875,
0.00847625732421875,
0.016876220703125,
0.01215362548828125,
0.0335693359375,
0.055511474609375,
-0.00920867919921875,
-0.0099334716796875,
0.0196685791015625,
0.05572509765625,
-0.046722412109375,
-0.056060791015625,
-0.03558349609375,
0.04595947265625,
-0.01003265380859375,
-0.05108642578125,
0.06927490234375,
0.048858642578125,
0.08599853515625,
-0.0322265625,
0.0579833984375,
0.0011386871337890625,
0.049072265625,
-0.02313232421875,
0.07147216796875,
-0.05316162109375,
0.0139312744140625,
-0.0225677490234375,
-0.07977294921875,
-0.0077056884765625,
0.053924560546875,
-0.038818359375,
0.02215576171875,
0.0487060546875,
0.058502197265625,
-0.00115203857421875,
0.0035114288330078125,
0.01059722900390625,
0.02032470703125,
-0.00466156005859375,
0.031951904296875,
0.0487060546875,
-0.058197021484375,
0.055419921875,
-0.064208984375,
-0.006725311279296875,
-0.0202178955078125,
-0.042938232421875,
-0.08056640625,
-0.03094482421875,
-0.032135009765625,
-0.045806884765625,
-0.00849151611328125,
0.06683349609375,
0.058319091796875,
-0.072998046875,
-0.0261383056640625,
-0.00311279296875,
-0.0238800048828125,
-0.0301666259765625,
-0.021392822265625,
0.03265380859375,
-0.0369873046875,
-0.0694580078125,
-0.00762176513671875,
-0.023223876953125,
-0.0032787322998046875,
-0.034027099609375,
-0.0023555755615234375,
-0.002918243408203125,
-0.005207061767578125,
0.0384521484375,
0.01153564453125,
-0.05096435546875,
-0.0091705322265625,
0.02056884765625,
-0.011749267578125,
0.0200347900390625,
0.02496337890625,
-0.044097900390625,
0.042205810546875,
0.03448486328125,
0.0248870849609375,
0.0394287109375,
0.00008183717727661133,
0.03240966796875,
-0.056671142578125,
0.015350341796875,
0.0280914306640625,
0.0367431640625,
0.020782470703125,
-0.03631591796875,
0.0211944580078125,
0.0139007568359375,
-0.058807373046875,
-0.039337158203125,
-0.0011577606201171875,
-0.08563232421875,
-0.007266998291015625,
0.093994140625,
-0.000885009765625,
-0.0133514404296875,
0.0120849609375,
-0.0220794677734375,
0.005565643310546875,
-0.03131103515625,
0.07427978515625,
0.0596923828125,
-0.00037860870361328125,
0.0004482269287109375,
-0.0341796875,
0.04632568359375,
0.049713134765625,
-0.0343017578125,
-0.0162200927734375,
0.035186767578125,
0.039398193359375,
0.01678466796875,
0.0263214111328125,
-0.011383056640625,
0.029052734375,
-0.015106201171875,
0.03167724609375,
-0.0078277587890625,
-0.01212310791015625,
-0.041290283203125,
0.00199127197265625,
0.0027027130126953125,
-0.031890869140625
]
] |
zarakiquemparte/kuchiki-1.1-l2-7b | 2023-09-16T00:33:38.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama2",
"license:other",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | zarakiquemparte | null | null | zarakiquemparte/kuchiki-1.1-l2-7b | 3 | 7,426 | transformers | 2023-09-15T21:47:46 | ---
license: other
tags:
- llama2
---
# Model Card: Kuchiki 1.1 L2 7b
This model uses [Nous Hermes Llama2 7b](https://huggingface.co/NousResearch/Nous-Hermes-llama-2-7b) (70%) as a base with [Airoboros L2 7B GPT4 2.0](https://huggingface.co/jondurbin/airoboros-l2-7b-gpt4-2.0) (30%) and the result of this merge was merged with [LimaRP Llama2 v2 7B Lora](https://huggingface.co/lemonilia/limarp-llama2-v2).
This merge of models(hermes and airoboros) was done with this [script](https://github.com/zarakiquemparte/zaraki-tools/blob/main/merge-cli.py)
This merge of Lora with Model was done with this [script](https://github.com/zarakiquemparte/zaraki-tools/blob/main/apply-lora.py)
Merge illustration:
## Usage:
Since this is a merge between Nous Hermes, Airoboros and LimaRP, the following instruction formats should work:
Alpaca 2:
```
### Instruction:
<prompt>
### Response:
<leave a newline blank for model to respond>
```
Alpaca LimaRP:
```
### 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}
```
## Bias, Risks, and Limitations
This model is not intended for supplying factual information or advice in any form
## Training Details
This model is merged and can be reproduced using the tools mentioned above. Please refer to all provided links for extra model-specific details. | 1,700 | [
[
-0.03350830078125,
-0.042266845703125,
0.029937744140625,
0.0230560302734375,
-0.043548583984375,
-0.020233154296875,
0.0201263427734375,
-0.054779052734375,
0.037567138671875,
0.07061767578125,
-0.059112548828125,
-0.029876708984375,
-0.043609619140625,
-0.0102691650390625,
-0.0260772705078125,
0.097412109375,
0.0028553009033203125,
-0.020538330078125,
0.00440216064453125,
-0.018524169921875,
-0.035888671875,
-0.035125732421875,
-0.06365966796875,
-0.0389404296875,
0.045013427734375,
0.035552978515625,
0.057403564453125,
0.0232086181640625,
0.03271484375,
0.026123046875,
-0.0285797119140625,
0.0188446044921875,
-0.0289154052734375,
0.0015697479248046875,
0.008544921875,
-0.040802001953125,
-0.0830078125,
0.0093536376953125,
0.033660888671875,
0.03265380859375,
-0.0220947265625,
0.013275146484375,
-0.01045989990234375,
0.0156707763671875,
-0.0292510986328125,
0.002010345458984375,
-0.0182952880859375,
0.01274871826171875,
0.0011968612670898438,
0.000003337860107421875,
-0.01515960693359375,
-0.0237884521484375,
-0.0012006759643554688,
-0.065185546875,
-0.004421234130859375,
-0.01296234130859375,
0.080810546875,
0.0173187255859375,
-0.04296875,
-0.016357421875,
-0.03936767578125,
0.042816162109375,
-0.08197021484375,
0.003467559814453125,
0.032501220703125,
0.0277557373046875,
-0.021453857421875,
-0.037750244140625,
-0.0584716796875,
-0.004833221435546875,
-0.0005216598510742188,
0.009765625,
-0.027740478515625,
-0.0190582275390625,
0.0318603515625,
0.02435302734375,
-0.0186309814453125,
0.0248870849609375,
-0.06396484375,
0.0020503997802734375,
0.0487060546875,
0.0265655517578125,
0.01244354248046875,
-0.0174407958984375,
-0.03692626953125,
-0.0301513671875,
-0.02996826171875,
0.004978179931640625,
0.04864501953125,
0.01611328125,
-0.06475830078125,
0.07342529296875,
-0.00946807861328125,
0.048248291015625,
0.0234832763671875,
-0.03436279296875,
0.03759765625,
-0.00325775146484375,
-0.0293121337890625,
-0.00494384765625,
0.0634765625,
0.042877197265625,
-0.0106353759765625,
0.0197906494140625,
0.0015783309936523438,
0.00640106201171875,
0.006069183349609375,
-0.059051513671875,
0.016876220703125,
0.031951904296875,
-0.0276031494140625,
-0.05584716796875,
-0.00913238525390625,
-0.0499267578125,
-0.0285491943359375,
0.0031070709228515625,
0.0131378173828125,
-0.020263671875,
-0.0263824462890625,
0.0096588134765625,
-0.006633758544921875,
0.05059814453125,
0.03485107421875,
-0.05767822265625,
0.02691650390625,
0.033355712890625,
0.03912353515625,
0.02105712890625,
-0.023040771484375,
-0.030364990234375,
0.01351165771484375,
-0.033111572265625,
0.055328369140625,
-0.00827789306640625,
-0.043670654296875,
-0.01381683349609375,
0.007495880126953125,
0.0108642578125,
-0.03289794921875,
0.051788330078125,
-0.0312347412109375,
0.02862548828125,
-0.015869140625,
-0.0241241455078125,
-0.0295257568359375,
0.0169830322265625,
-0.061309814453125,
0.07244873046875,
0.02337646484375,
-0.055999755859375,
-0.0023212432861328125,
-0.0626220703125,
-0.0250091552734375,
-0.0103302001953125,
0.0098419189453125,
-0.04486083984375,
-0.003559112548828125,
0.0008859634399414062,
0.030059814453125,
-0.022552490234375,
-0.0097198486328125,
-0.046539306640625,
-0.0207977294921875,
0.019989013671875,
0.0048675537109375,
0.06787109375,
0.020965576171875,
-0.0085296630859375,
-0.004375457763671875,
-0.061248779296875,
0.006053924560546875,
0.032440185546875,
-0.0268096923828125,
-0.0009670257568359375,
-0.02154541015625,
0.014251708984375,
0.00848388671875,
0.029205322265625,
-0.0221710205078125,
0.049835205078125,
-0.02325439453125,
0.01904296875,
0.0374755859375,
0.0025997161865234375,
0.021820068359375,
-0.041107177734375,
0.0295867919921875,
0.0114593505859375,
0.0200958251953125,
-0.005962371826171875,
-0.0635986328125,
-0.080322265625,
-0.0111541748046875,
-0.00090789794921875,
0.035308837890625,
-0.0236358642578125,
0.048004150390625,
-0.0015926361083984375,
-0.060211181640625,
-0.0343017578125,
-0.001720428466796875,
0.01068115234375,
0.038421630859375,
0.01293182373046875,
-0.0249786376953125,
-0.053924560546875,
-0.06475830078125,
0.0220184326171875,
-0.029571533203125,
0.00075531005859375,
0.0166473388671875,
0.039825439453125,
-0.0433349609375,
0.045867919921875,
-0.042327880859375,
-0.02044677734375,
-0.040374755859375,
0.016876220703125,
0.03955078125,
0.050140380859375,
0.05157470703125,
-0.0438232421875,
-0.02685546875,
0.007781982421875,
-0.06365966796875,
-0.006267547607421875,
0.01438140869140625,
-0.01007843017578125,
0.0209197998046875,
0.0041656494140625,
-0.07293701171875,
0.0537109375,
0.0469970703125,
-0.034576416015625,
0.0147247314453125,
-0.022491455078125,
0.0200958251953125,
-0.10833740234375,
0.0297393798828125,
-0.022613525390625,
0.0008921623229980469,
-0.055206298828125,
0.025848388671875,
-0.0139617919921875,
-0.017181396484375,
-0.04302978515625,
0.06756591796875,
-0.0265960693359375,
-0.01274871826171875,
-0.03558349609375,
0.005809783935546875,
0.0144195556640625,
0.03607177734375,
0.0010290145874023438,
0.01284027099609375,
0.0277862548828125,
-0.043609619140625,
0.0452880859375,
0.04949951171875,
-0.01284027099609375,
0.043792724609375,
-0.05413818359375,
0.0284576416015625,
0.0012044906616210938,
0.0350341796875,
-0.0706787109375,
-0.0322265625,
0.058013916015625,
-0.03216552734375,
0.002933502197265625,
-0.0136260986328125,
-0.03753662109375,
-0.030731201171875,
-0.0196685791015625,
0.0265655517578125,
0.05633544921875,
-0.034332275390625,
0.06671142578125,
0.0038318634033203125,
0.006595611572265625,
-0.0462646484375,
-0.05670166015625,
-0.0187835693359375,
-0.029205322265625,
-0.051300048828125,
0.014862060546875,
-0.022705078125,
-0.007717132568359375,
0.0041961669921875,
0.0034046173095703125,
-0.0255279541015625,
-0.003482818603515625,
0.031829833984375,
0.050140380859375,
-0.0192718505859375,
-0.035552978515625,
0.014892578125,
-0.0023097991943359375,
-0.0274200439453125,
0.03143310546875,
0.053131103515625,
0.0028533935546875,
-0.0290069580078125,
-0.0523681640625,
0.02032470703125,
0.0721435546875,
-0.0115509033203125,
0.06671142578125,
0.04638671875,
-0.0208587646484375,
0.023040771484375,
-0.06427001953125,
0.00360107421875,
-0.0279998779296875,
0.0117645263671875,
-0.0286712646484375,
-0.033782958984375,
0.0687255859375,
0.02191162109375,
0.0084686279296875,
0.041839599609375,
0.02734375,
-0.00928497314453125,
0.06561279296875,
0.059600830078125,
-0.0207366943359375,
0.017822265625,
-0.0478515625,
0.0173187255859375,
-0.08160400390625,
-0.038482666015625,
-0.0259246826171875,
-0.022705078125,
-0.0391845703125,
-0.03338623046875,
0.008544921875,
0.0306549072265625,
-0.01348876953125,
0.05413818359375,
-0.01259613037109375,
0.0206298828125,
0.019866943359375,
0.006488800048828125,
0.02685546875,
0.00832366943359375,
0.0142059326171875,
0.01486968994140625,
-0.03851318359375,
-0.0273895263671875,
0.08135986328125,
0.0318603515625,
0.060211181640625,
0.02276611328125,
0.05975341796875,
-0.0012912750244140625,
0.0164642333984375,
-0.045440673828125,
0.04833984375,
0.004657745361328125,
-0.044342041015625,
-0.01519775390625,
-0.01541900634765625,
-0.072021484375,
0.0250091552734375,
-0.0206298828125,
-0.05438232421875,
0.008148193359375,
0.0010213851928710938,
-0.046478271484375,
0.0232086181640625,
-0.038421630859375,
0.05120849609375,
-0.008575439453125,
-0.0058135986328125,
-0.016815185546875,
-0.04412841796875,
0.0455322265625,
0.0030574798583984375,
0.00032711029052734375,
-0.0157012939453125,
-0.01404571533203125,
0.059661865234375,
-0.043975830078125,
0.06597900390625,
0.004596710205078125,
-0.0277252197265625,
0.0288848876953125,
-0.010528564453125,
0.040679931640625,
-0.01131439208984375,
0.00457000732421875,
0.017333984375,
-0.0029926300048828125,
-0.0223388671875,
-0.044769287109375,
0.0579833984375,
-0.081787109375,
-0.0264892578125,
-0.03387451171875,
-0.0280914306640625,
0.007755279541015625,
0.00670623779296875,
0.033966064453125,
0.0160064697265625,
-0.00995635986328125,
-0.00696563720703125,
0.03546142578125,
-0.01537322998046875,
0.0194244384765625,
0.06683349609375,
-0.0276641845703125,
-0.044891357421875,
0.02935791015625,
-0.01934814453125,
0.01456451416015625,
0.008270263671875,
0.0039825439453125,
-0.0141143798828125,
0.006122589111328125,
-0.033050537109375,
0.055206298828125,
-0.0579833984375,
-0.0166473388671875,
-0.0323486328125,
-0.03204345703125,
-0.04278564453125,
-0.00353240966796875,
-0.0260009765625,
-0.02947998046875,
-0.0265655517578125,
-0.003021240234375,
0.03375244140625,
0.0645751953125,
-0.01119232177734375,
0.05426025390625,
-0.051605224609375,
0.032989501953125,
0.021575927734375,
0.0034122467041015625,
0.003376007080078125,
-0.060791015625,
0.00452423095703125,
0.008544921875,
-0.0305938720703125,
-0.08026123046875,
0.0438232421875,
-0.011810302734375,
0.041748046875,
0.0355224609375,
-0.0099945068359375,
0.06103515625,
-0.0086517333984375,
0.047698974609375,
0.0261383056640625,
-0.055450439453125,
0.05328369140625,
-0.032562255859375,
0.014068603515625,
0.01103973388671875,
0.030426025390625,
-0.052581787109375,
-0.01554107666015625,
-0.06842041015625,
-0.043731689453125,
0.0760498046875,
0.0234527587890625,
0.00896453857421875,
0.0225677490234375,
0.0217742919921875,
-0.01561737060546875,
0.020843505859375,
-0.072509765625,
-0.031494140625,
-0.006717681884765625,
0.003093719482421875,
-0.004970550537109375,
-0.02203369140625,
-0.0256805419921875,
-0.01308441162109375,
0.062744140625,
0.0011472702026367188,
0.019256591796875,
0.0006771087646484375,
0.0204925537109375,
-0.0277099609375,
0.006725311279296875,
0.055206298828125,
0.0110626220703125,
-0.0295562744140625,
-0.0276031494140625,
0.0209503173828125,
-0.01515960693359375,
0.0090484619140625,
0.00897979736328125,
0.0014066696166992188,
-0.00844573974609375,
0.03955078125,
0.07525634765625,
0.025115966796875,
-0.032928466796875,
0.03271484375,
-0.002040863037109375,
-0.0179595947265625,
-0.01543426513671875,
0.007755279541015625,
0.017364501953125,
0.026885986328125,
-0.0019092559814453125,
0.01222991943359375,
-0.0066680908203125,
-0.0562744140625,
-0.025390625,
0.0234832763671875,
0.00214385986328125,
-0.017181396484375,
0.043792724609375,
0.0166473388671875,
-0.0195465087890625,
0.055908203125,
0.002574920654296875,
-0.0245819091796875,
0.056915283203125,
0.041046142578125,
0.0479736328125,
-0.035491943359375,
0.0004546642303466797,
0.042388916015625,
0.005298614501953125,
-0.006420135498046875,
0.0187835693359375,
-0.005096435546875,
-0.05523681640625,
-0.01418304443359375,
-0.032073974609375,
-0.0269317626953125,
0.05126953125,
-0.047576904296875,
0.0484619140625,
-0.036590576171875,
-0.015899658203125,
-0.006786346435546875,
0.01221466064453125,
-0.03460693359375,
0.007007598876953125,
0.01495361328125,
0.05975341796875,
-0.0806884765625,
0.0650634765625,
0.04949951171875,
-0.05242919921875,
-0.08319091796875,
-0.0200653076171875,
-0.006961822509765625,
-0.076171875,
0.05352783203125,
0.00701904296875,
0.00714111328125,
-0.0310211181640625,
-0.044891357421875,
-0.07427978515625,
0.09356689453125,
0.02911376953125,
-0.0179443359375,
-0.0060272216796875,
0.001064300537109375,
0.03216552734375,
-0.0360107421875,
0.03515625,
0.022064208984375,
0.03228759765625,
0.043670654296875,
-0.08624267578125,
-0.00516510009765625,
-0.028564453125,
-0.0027713775634765625,
-0.006107330322265625,
-0.06817626953125,
0.08294677734375,
-0.029296875,
0.0053253173828125,
0.06451416015625,
0.04736328125,
0.041534423828125,
0.0026607513427734375,
0.034393310546875,
0.07147216796875,
0.040557861328125,
0.0113067626953125,
0.06109619140625,
-0.014251708984375,
0.0305023193359375,
0.0787353515625,
-0.037841796875,
0.0770263671875,
0.02691650390625,
0.00030732154846191406,
0.06109619140625,
0.040252685546875,
-0.01203155517578125,
0.0301666259765625,
-0.0037994384765625,
-0.01397705078125,
-0.007099151611328125,
0.004512786865234375,
-0.058258056640625,
0.046600341796875,
0.0088348388671875,
-0.019989013671875,
-0.0115509033203125,
-0.0216217041015625,
0.00893402099609375,
-0.0233917236328125,
-0.007076263427734375,
0.0450439453125,
-0.005153656005859375,
-0.06536865234375,
0.04705810546875,
0.0154876708984375,
0.045867919921875,
-0.07806396484375,
-0.02642822265625,
-0.0455322265625,
0.019012451171875,
-0.0257110595703125,
-0.03912353515625,
0.008453369140625,
-0.00231170654296875,
-0.01371002197265625,
0.018310546875,
0.042144775390625,
-0.0261688232421875,
-0.046142578125,
0.02947998046875,
0.0203399658203125,
0.03271484375,
0.0246429443359375,
-0.0499267578125,
0.035980224609375,
0.0020885467529296875,
-0.00818634033203125,
0.0279388427734375,
0.010101318359375,
0.005584716796875,
0.06463623046875,
0.046478271484375,
-0.00325775146484375,
-0.01511383056640625,
-0.0013818740844726562,
0.06927490234375,
-0.028594970703125,
-0.03924560546875,
-0.041717529296875,
0.044342041015625,
-0.00785064697265625,
-0.0270843505859375,
0.049896240234375,
0.051055908203125,
0.039947509765625,
-0.0178985595703125,
0.0482177734375,
-0.0137481689453125,
0.035858154296875,
-0.03594970703125,
0.0528564453125,
-0.052001953125,
0.0182342529296875,
-0.0238800048828125,
-0.07135009765625,
0.01488494873046875,
0.06732177734375,
0.00955963134765625,
0.00582122802734375,
0.03509521484375,
0.055267333984375,
-0.015777587890625,
-0.0167236328125,
0.024658203125,
0.018524169921875,
0.01073455810546875,
0.052154541015625,
0.06768798828125,
-0.0738525390625,
0.0318603515625,
-0.019989013671875,
-0.0155792236328125,
-0.022857666015625,
-0.0648193359375,
-0.0714111328125,
-0.0241241455078125,
-0.024139404296875,
-0.032012939453125,
-0.0172271728515625,
0.05865478515625,
0.0672607421875,
-0.03973388671875,
-0.04913330078125,
0.019561767578125,
-0.00493621826171875,
-0.0018320083618164062,
-0.01248931884765625,
0.0020771026611328125,
0.012115478515625,
-0.060760498046875,
0.0230865478515625,
-0.002063751220703125,
0.048309326171875,
-0.012603759765625,
-0.02935791015625,
-0.0058135986328125,
0.00567626953125,
0.0283660888671875,
0.03863525390625,
-0.07452392578125,
0.00629425048828125,
-0.0117340087890625,
-0.004039764404296875,
-0.0014820098876953125,
0.033843994140625,
-0.0465087890625,
-0.00571441650390625,
0.015380859375,
0.002071380615234375,
0.046844482421875,
-0.01031494140625,
0.02606201171875,
-0.037841796875,
0.0305023193359375,
-0.0135345458984375,
0.036834716796875,
0.0256805419921875,
-0.0277862548828125,
0.047210693359375,
0.01470184326171875,
-0.0224456787109375,
-0.061004638671875,
0.003326416015625,
-0.11297607421875,
-0.0046844482421875,
0.0748291015625,
0.0019626617431640625,
-0.02001953125,
0.036346435546875,
-0.04534912109375,
0.017578125,
-0.030120849609375,
0.0308685302734375,
0.04296875,
-0.0235748291015625,
-0.005527496337890625,
-0.0098724365234375,
0.0108642578125,
0.009918212890625,
-0.06494140625,
-0.012176513671875,
0.0205535888671875,
0.039703369140625,
0.036712646484375,
0.061614990234375,
0.0018396377563476562,
0.0240478515625,
0.000568389892578125,
0.0230865478515625,
-0.00860595703125,
-0.00004851818084716797,
-0.0164947509765625,
-0.01522064208984375,
-0.0013513565063476562,
-0.00952911376953125
]
] |
allegro/herbert-base-cased | 2022-06-09T11:36:39.000Z | [
"transformers",
"pytorch",
"tf",
"jax",
"bert",
"feature-extraction",
"herbert",
"pl",
"license:cc-by-4.0",
"endpoints_compatible",
"region:us"
] | feature-extraction | allegro | null | null | allegro/herbert-base-cased | 10 | 7,425 | transformers | 2022-03-02T23:29:05 | ---
language: pl
tags:
- herbert
license: cc-by-4.0
---
# HerBERT
**[HerBERT](https://en.wikipedia.org/wiki/Zbigniew_Herbert)** is a BERT-based Language Model trained on Polish corpora
using Masked Language Modelling (MLM) and Sentence Structural Objective (SSO) with dynamic masking of whole words. For more details, please refer to: [HerBERT: Efficiently Pretrained Transformer-based Language Model for Polish](https://www.aclweb.org/anthology/2021.bsnlp-1.1/).
Model training and experiments were conducted with [transformers](https://github.com/huggingface/transformers) in version 2.9.
## Corpus
HerBERT was trained on six different corpora available for Polish language:
| Corpus | Tokens | Documents |
| :------ | ------: | ------: |
| [CCNet Middle](https://github.com/facebookresearch/cc_net) | 3243M | 7.9M |
| [CCNet Head](https://github.com/facebookresearch/cc_net) | 2641M | 7.0M |
| [National Corpus of Polish](http://nkjp.pl/index.php?page=14&lang=1)| 1357M | 3.9M |
| [Open Subtitles](http://opus.nlpl.eu/OpenSubtitles-v2018.php) | 1056M | 1.1M
| [Wikipedia](https://dumps.wikimedia.org/) | 260M | 1.4M |
| [Wolne Lektury](https://wolnelektury.pl/) | 41M | 5.5k |
## Tokenizer
The training dataset was tokenized into subwords using a character level byte-pair encoding (``CharBPETokenizer``) with
a vocabulary size of 50k tokens. The tokenizer itself was trained with a [tokenizers](https://github.com/huggingface/tokenizers) library.
We kindly encourage you to use the ``Fast`` version of the tokenizer, namely ``HerbertTokenizerFast``.
## Usage
Example code:
```python
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("allegro/herbert-base-cased")
model = AutoModel.from_pretrained("allegro/herbert-base-cased")
output = model(
**tokenizer.batch_encode_plus(
[
(
"A potem szedł środkiem drogi w kurzawie, bo zamiatał nogami, ślepy dziad prowadzony przez tłustego kundla na sznurku.",
"A potem leciał od lasu chłopak z butelką, ale ten ujrzawszy księdza przy drodze okrążył go z dala i biegł na przełaj pól do karczmy."
)
],
padding='longest',
add_special_tokens=True,
return_tensors='pt'
)
)
```
## License
CC BY 4.0
## Citation
If you use this model, please cite the following paper:
```
@inproceedings{mroczkowski-etal-2021-herbert,
title = "{H}er{BERT}: Efficiently Pretrained Transformer-based Language Model for {P}olish",
author = "Mroczkowski, Robert and
Rybak, Piotr and
Wr{\\'o}blewska, Alina and
Gawlik, Ireneusz",
booktitle = "Proceedings of the 8th Workshop on Balto-Slavic Natural Language Processing",
month = apr,
year = "2021",
address = "Kiyv, Ukraine",
publisher = "Association for Computational Linguistics",
url = "https://www.aclweb.org/anthology/2021.bsnlp-1.1",
pages = "1--10",
}
```
## Authors
The model was trained by [**Machine Learning Research Team at Allegro**](https://ml.allegro.tech/) and [**Linguistic Engineering Group at Institute of Computer Science, Polish Academy of Sciences**](http://zil.ipipan.waw.pl/).
You can contact us at: <a href="mailto:klejbenchmark@allegro.pl">klejbenchmark@allegro.pl</a> | 3,271 | [
[
-0.0318603515625,
-0.057647705078125,
0.0200042724609375,
0.016357421875,
-0.0217132568359375,
-0.002376556396484375,
-0.03997802734375,
-0.03240966796875,
0.0184326171875,
0.0272979736328125,
-0.054779052734375,
-0.043243408203125,
-0.050506591796875,
0.0087432861328125,
-0.019989013671875,
0.082763671875,
-0.01438140869140625,
0.024322509765625,
0.0116424560546875,
0.003021240234375,
0.007022857666015625,
-0.053955078125,
-0.019012451171875,
-0.021026611328125,
0.017608642578125,
0.0018138885498046875,
0.03790283203125,
0.02606201171875,
0.0335693359375,
0.027862548828125,
-0.00872039794921875,
-0.00124359130859375,
-0.025726318359375,
-0.006103515625,
0.0120697021484375,
-0.0234527587890625,
-0.02825927734375,
-0.0011816024780273438,
0.05194091796875,
0.051177978515625,
-0.0137481689453125,
0.01180267333984375,
-0.01336669921875,
0.039154052734375,
-0.032470703125,
0.0173492431640625,
-0.053497314453125,
0.006481170654296875,
-0.02667236328125,
0.0178985595703125,
-0.050506591796875,
-0.00641632080078125,
0.0005865097045898438,
-0.04241943359375,
0.02978515625,
0.01364898681640625,
0.10870361328125,
0.0114593505859375,
-0.00177764892578125,
-0.0183258056640625,
-0.04107666015625,
0.06341552734375,
-0.06591796875,
0.034698486328125,
0.018707275390625,
0.007747650146484375,
-0.01611328125,
-0.06085205078125,
-0.0521240234375,
-0.0254058837890625,
-0.01352691650390625,
0.019683837890625,
-0.02227783203125,
0.0096893310546875,
0.021728515625,
0.030487060546875,
-0.045684814453125,
0.00011998414993286133,
-0.03143310546875,
-0.03076171875,
0.041656494140625,
-0.01168060302734375,
0.032073974609375,
-0.017822265625,
-0.03240966796875,
-0.0271148681640625,
-0.0310516357421875,
0.0028076171875,
0.0338134765625,
0.0295867919921875,
-0.015228271484375,
0.0382080078125,
-0.0017442703247070312,
0.05487060546875,
0.01450347900390625,
-0.0038471221923828125,
0.053802490234375,
-0.0144805908203125,
-0.0120086669921875,
0.004558563232421875,
0.09130859375,
-0.006412506103515625,
0.027496337890625,
-0.0204925537109375,
-0.015655517578125,
0.0074462890625,
0.0142669677734375,
-0.05633544921875,
-0.02032470703125,
0.0008254051208496094,
-0.0302734375,
-0.01085662841796875,
0.026641845703125,
-0.03656005859375,
0.01412200927734375,
-0.0128173828125,
0.0286712646484375,
-0.060882568359375,
-0.019378662109375,
0.01161956787109375,
0.0045318603515625,
0.0220184326171875,
-0.00475311279296875,
-0.07562255859375,
0.0181121826171875,
0.034149169921875,
0.0423583984375,
-0.00763702392578125,
-0.039093017578125,
-0.031280517578125,
-0.01224517822265625,
-0.02154541015625,
0.03662109375,
-0.0156707763671875,
-0.008087158203125,
0.005275726318359375,
0.01433563232421875,
-0.0197906494140625,
-0.024017333984375,
0.03717041015625,
-0.03363037109375,
0.03497314453125,
0.0074920654296875,
-0.046630859375,
-0.0026702880859375,
0.000713348388671875,
-0.03155517578125,
0.09991455078125,
0.0305633544921875,
-0.06268310546875,
0.028594970703125,
-0.04132080078125,
-0.032989501953125,
0.0001938343048095703,
0.0024051666259765625,
-0.046966552734375,
0.0124053955078125,
0.02471923828125,
0.022125244140625,
-0.005443572998046875,
0.036590576171875,
-0.016448974609375,
-0.0137786865234375,
0.018402099609375,
-0.01502227783203125,
0.07940673828125,
0.007801055908203125,
-0.0364990234375,
0.02630615234375,
-0.07208251953125,
0.0017347335815429688,
0.01457977294921875,
-0.026123046875,
-0.0160980224609375,
-0.0221405029296875,
0.0201568603515625,
0.03314208984375,
0.02606201171875,
-0.06494140625,
0.01544189453125,
-0.060089111328125,
0.026763916015625,
0.054718017578125,
-0.026611328125,
0.0299072265625,
-0.0022945404052734375,
0.03076171875,
0.0015811920166015625,
0.0240478515625,
-0.00003355741500854492,
-0.04144287109375,
-0.0643310546875,
-0.032012939453125,
0.039703369140625,
0.0413818359375,
-0.06121826171875,
0.053955078125,
-0.02099609375,
-0.04498291015625,
-0.050750732421875,
-0.0016183853149414062,
0.033477783203125,
0.039703369140625,
0.0406494140625,
-0.024810791015625,
-0.05426025390625,
-0.08099365234375,
-0.00982666015625,
-0.0221405029296875,
-0.007610321044921875,
0.0087432861328125,
0.04449462890625,
-0.0222625732421875,
0.08111572265625,
-0.0208282470703125,
-0.014862060546875,
-0.027984619140625,
0.0141143798828125,
0.03839111328125,
0.040557861328125,
0.043792724609375,
-0.047698974609375,
-0.051727294921875,
-0.0176544189453125,
-0.050750732421875,
-0.014434814453125,
0.002262115478515625,
-0.022216796875,
0.0433349609375,
0.033538818359375,
-0.054779052734375,
0.016265869140625,
0.03961181640625,
-0.03363037109375,
0.05218505859375,
-0.0274658203125,
-0.01788330078125,
-0.082275390625,
0.007610321044921875,
-0.0127105712890625,
-0.0059814453125,
-0.05426025390625,
-0.0017709732055664062,
0.009674072265625,
-0.007266998291015625,
-0.05010986328125,
0.046112060546875,
-0.03729248046875,
-0.006946563720703125,
0.000047147274017333984,
0.00994110107421875,
-0.0204925537109375,
0.0648193359375,
0.0151214599609375,
0.051513671875,
0.0557861328125,
-0.0294189453125,
0.0157318115234375,
0.041168212890625,
-0.031524658203125,
0.01297760009765625,
-0.04962158203125,
0.0019474029541015625,
-0.008026123046875,
0.016998291015625,
-0.058685302734375,
-0.0012950897216796875,
0.031341552734375,
-0.042510986328125,
0.044952392578125,
-0.0017833709716796875,
-0.052337646484375,
-0.0447998046875,
-0.01702880859375,
0.002166748046875,
0.050628662109375,
-0.040252685546875,
0.060699462890625,
0.022064208984375,
-0.00548553466796875,
-0.053924560546875,
-0.058258056640625,
-0.0005726814270019531,
-0.0029544830322265625,
-0.0557861328125,
0.047698974609375,
-0.012969970703125,
-0.00350189208984375,
0.006256103515625,
0.00039839744567871094,
-0.0025997161865234375,
-0.0031986236572265625,
0.0169830322265625,
0.043609619140625,
-0.015869140625,
0.0022296905517578125,
-0.0030059814453125,
-0.02154541015625,
-0.00530242919921875,
-0.01428985595703125,
0.062103271484375,
-0.035888671875,
-0.006206512451171875,
-0.036407470703125,
0.0107879638671875,
0.0404052734375,
-0.01477813720703125,
0.0738525390625,
0.0684814453125,
-0.01239776611328125,
-0.0003809928894042969,
-0.039398193359375,
-0.0213470458984375,
-0.0343017578125,
0.0290679931640625,
-0.03076171875,
-0.06463623046875,
0.04052734375,
0.008148193359375,
0.002567291259765625,
0.057373046875,
0.0587158203125,
0.0046844482421875,
0.06671142578125,
0.052337646484375,
-0.0183563232421875,
0.04876708984375,
-0.03466796875,
0.020660400390625,
-0.06292724609375,
-0.00995635986328125,
-0.0299835205078125,
0.0032196044921875,
-0.058837890625,
-0.016754150390625,
0.00724029541015625,
0.01129913330078125,
-0.0263824462890625,
0.054168701171875,
-0.038299560546875,
0.0007719993591308594,
0.0501708984375,
-0.003875732421875,
-0.00048065185546875,
0.0107269287109375,
-0.036407470703125,
-0.0030040740966796875,
-0.0648193359375,
-0.036712646484375,
0.07427978515625,
0.0294647216796875,
0.0474853515625,
-0.0090484619140625,
0.067138671875,
-0.0020904541015625,
0.0244293212890625,
-0.053070068359375,
0.04632568359375,
-0.0143280029296875,
-0.05670166015625,
-0.03436279296875,
-0.033966064453125,
-0.06671142578125,
0.0279541015625,
-0.0227813720703125,
-0.056243896484375,
0.0183563232421875,
0.0009226799011230469,
-0.00838470458984375,
0.02374267578125,
-0.042694091796875,
0.071533203125,
-0.0142669677734375,
-0.0101776123046875,
-0.0020313262939453125,
-0.06378173828125,
0.00759124755859375,
-0.00039124488830566406,
0.0233917236328125,
-0.00539398193359375,
0.01500701904296875,
0.06353759765625,
-0.0477294921875,
0.053863525390625,
-0.0163116455078125,
-0.0029048919677734375,
0.017242431640625,
-0.014007568359375,
0.034393310546875,
-0.020416259765625,
-0.0063629150390625,
0.04107666015625,
-0.009307861328125,
-0.032501220703125,
-0.0240478515625,
0.033935546875,
-0.0723876953125,
-0.0232696533203125,
-0.04010009765625,
-0.040863037109375,
-0.0126800537109375,
0.028900146484375,
0.045562744140625,
0.027130126953125,
-0.006763458251953125,
0.00891876220703125,
0.0394287109375,
-0.03594970703125,
0.042022705078125,
0.051300048828125,
-0.011962890625,
-0.0273284912109375,
0.06378173828125,
0.0057525634765625,
0.0158843994140625,
0.0216217041015625,
0.01513671875,
-0.0101776123046875,
-0.032135009765625,
-0.042694091796875,
0.0509033203125,
-0.047882080078125,
0.0001876354217529297,
-0.052703857421875,
-0.024688720703125,
-0.03802490234375,
-0.0019083023071289062,
-0.040496826171875,
-0.034027099609375,
-0.0095672607421875,
-0.007175445556640625,
0.0163116455078125,
0.0294036865234375,
-0.007282257080078125,
0.022125244140625,
-0.04522705078125,
0.0151214599609375,
0.00021970272064208984,
0.0194549560546875,
-0.0162506103515625,
-0.055328369140625,
-0.0229644775390625,
0.00002467632293701172,
-0.0118255615234375,
-0.047943115234375,
0.04937744140625,
0.01207733154296875,
0.05023193359375,
0.0112457275390625,
0.0161590576171875,
0.035552978515625,
-0.054229736328125,
0.06683349609375,
0.019317626953125,
-0.075927734375,
0.039947509765625,
-0.01263427734375,
0.0168304443359375,
0.049285888671875,
0.02532958984375,
-0.0287017822265625,
-0.0528564453125,
-0.07373046875,
-0.08575439453125,
0.07940673828125,
0.0254974365234375,
0.0211029052734375,
-0.00940704345703125,
0.0114593505859375,
-0.0017347335815429688,
0.017730712890625,
-0.07037353515625,
-0.033477783203125,
-0.0217742919921875,
-0.02532958984375,
-0.005420684814453125,
-0.02447509765625,
0.01006317138671875,
-0.0312042236328125,
0.09625244140625,
0.007244110107421875,
0.031829833984375,
0.020660400390625,
-0.035858154296875,
-0.0024700164794921875,
0.00995635986328125,
0.054718017578125,
0.03936767578125,
-0.0239715576171875,
-0.0105133056640625,
0.019378662109375,
-0.04522705078125,
-0.00881195068359375,
0.0174102783203125,
-0.0238037109375,
0.03106689453125,
0.03619384765625,
0.093994140625,
0.0181121826171875,
-0.049591064453125,
0.045318603515625,
-0.0023517608642578125,
-0.033416748046875,
-0.03802490234375,
-0.01422882080078125,
-0.0078125,
-0.006805419921875,
0.0300445556640625,
-0.0121307373046875,
-0.01161956787109375,
-0.040283203125,
0.017059326171875,
0.0180206298828125,
-0.044647216796875,
-0.01526641845703125,
0.040985107421875,
-0.00716400146484375,
-0.021881103515625,
0.064208984375,
-0.004016876220703125,
-0.056396484375,
0.0350341796875,
0.0450439453125,
0.059417724609375,
-0.017486572265625,
0.0094757080078125,
0.043609619140625,
0.026458740234375,
-0.005218505859375,
0.0098419189453125,
-0.00673675537109375,
-0.06268310546875,
-0.0275421142578125,
-0.0712890625,
-0.0102691650390625,
0.0045166015625,
-0.048492431640625,
0.01528167724609375,
-0.015106201171875,
-0.0179290771484375,
0.0019502639770507812,
-0.007160186767578125,
-0.0306396484375,
0.00807952880859375,
0.01406097412109375,
0.067626953125,
-0.06121826171875,
0.0841064453125,
0.0411376953125,
-0.0462646484375,
-0.0677490234375,
-0.003246307373046875,
-0.030426025390625,
-0.05950927734375,
0.07183837890625,
0.008087158203125,
-0.00061798095703125,
-0.0023555755615234375,
-0.031158447265625,
-0.06463623046875,
0.053466796875,
0.033447265625,
-0.04815673828125,
-0.01180267333984375,
0.00978851318359375,
0.050445556640625,
-0.0148773193359375,
0.014892578125,
0.0298614501953125,
0.038299560546875,
-0.001697540283203125,
-0.0758056640625,
-0.004619598388671875,
-0.0360107421875,
0.01515960693359375,
0.0159454345703125,
-0.03271484375,
0.072265625,
0.003849029541015625,
-0.00777435302734375,
0.007625579833984375,
0.0555419921875,
0.01375579833984375,
-0.0061187744140625,
0.031707763671875,
0.045013427734375,
0.043365478515625,
-0.0179443359375,
0.087646484375,
-0.0556640625,
0.0560302734375,
0.07281494140625,
0.006641387939453125,
0.058441162109375,
0.04046630859375,
-0.0269012451171875,
0.06573486328125,
0.040771484375,
-0.017059326171875,
0.040679931640625,
0.0199127197265625,
-0.0069732666015625,
-0.0198822021484375,
0.0203857421875,
-0.0252838134765625,
0.025726318359375,
0.01232147216796875,
-0.038238525390625,
-0.01033782958984375,
0.0273895263671875,
0.00885772705078125,
0.001995086669921875,
-0.0035858154296875,
0.04949951171875,
0.005374908447265625,
-0.045654296875,
0.060546875,
0.01220703125,
0.06689453125,
-0.065185546875,
0.024810791015625,
-0.0017757415771484375,
0.01690673828125,
-0.005496978759765625,
-0.032958984375,
0.0189056396484375,
0.0003139972686767578,
-0.016143798828125,
-0.0275115966796875,
0.042449951171875,
-0.036712646484375,
-0.05462646484375,
0.02960205078125,
0.01605224609375,
0.02606201171875,
0.00775909423828125,
-0.07183837890625,
0.00357818603515625,
-0.01097869873046875,
-0.041778564453125,
0.023468017578125,
0.0200958251953125,
-0.004161834716796875,
0.035919189453125,
0.0496826171875,
0.0101776123046875,
0.01116180419921875,
0.0198974609375,
0.05596923828125,
-0.0270843505859375,
-0.042205810546875,
-0.07086181640625,
0.040374755859375,
-0.01024627685546875,
-0.017242431640625,
0.068603515625,
0.05218505859375,
0.07550048828125,
-0.0273284912109375,
0.04315185546875,
-0.019378662109375,
0.0305328369140625,
-0.031494140625,
0.056182861328125,
-0.0272369384765625,
-0.002490997314453125,
-0.032806396484375,
-0.06927490234375,
-0.0241546630859375,
0.07135009765625,
-0.036041259765625,
0.006496429443359375,
0.0474853515625,
0.055511474609375,
0.0070343017578125,
-0.0247039794921875,
0.018280029296875,
0.0355224609375,
0.010955810546875,
0.033538818359375,
0.039093017578125,
-0.04705810546875,
0.0389404296875,
-0.03704833984375,
-0.01560211181640625,
-0.018707275390625,
-0.049835205078125,
-0.06646728515625,
-0.0556640625,
-0.0245361328125,
-0.033477783203125,
0.0016355514526367188,
0.075927734375,
0.055816650390625,
-0.07275390625,
-0.0222625732421875,
-0.0252838134765625,
-0.01203155517578125,
-0.015960693359375,
-0.0165557861328125,
0.0458984375,
-0.042327880859375,
-0.05596923828125,
0.0202789306640625,
0.00933074951171875,
0.005313873291015625,
-0.0140838623046875,
-0.018798828125,
-0.027099609375,
0.004825592041015625,
0.044708251953125,
0.0117950439453125,
-0.06549072265625,
-0.00856781005859375,
0.01279449462890625,
-0.0185699462890625,
0.0195159912109375,
0.04119873046875,
-0.052459716796875,
0.030792236328125,
0.0292205810546875,
0.0197906494140625,
0.0716552734375,
-0.01265716552734375,
0.030731201171875,
-0.0465087890625,
0.030487060546875,
0.01287841796875,
0.038055419921875,
0.0299835205078125,
-0.018402099609375,
0.036712646484375,
0.0282440185546875,
-0.0232696533203125,
-0.06488037109375,
0.00516510009765625,
-0.0777587890625,
-0.033355712890625,
0.07781982421875,
-0.0214691162109375,
-0.03961181640625,
0.0085601806640625,
-0.025970458984375,
0.03656005859375,
-0.0239715576171875,
0.053466796875,
0.06494140625,
-0.0027370452880859375,
-0.0032062530517578125,
-0.0296783447265625,
0.033905029296875,
0.031402587890625,
-0.042236328125,
0.0037288665771484375,
0.018798828125,
0.029998779296875,
0.0305633544921875,
0.04656982421875,
-0.018890380859375,
0.00794219970703125,
-0.006320953369140625,
0.036590576171875,
-0.0164337158203125,
-0.01471710205078125,
-0.02392578125,
0.0021190643310546875,
-0.0120697021484375,
-0.0050811767578125
]
] |
timm/tf_efficientnet_b5.ns_jft_in1k | 2023-04-27T21:21:37.000Z | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"dataset:imagenet-1k",
"arxiv:1905.11946",
"arxiv:1911.04252",
"license:apache-2.0",
"region:us"
] | image-classification | timm | null | null | timm/tf_efficientnet_b5.ns_jft_in1k | 0 | 7,424 | timm | 2022-12-13T00:04:12 | ---
tags:
- image-classification
- timm
library_name: timm
license: apache-2.0
datasets:
- imagenet-1k
---
# Model card for tf_efficientnet_b5.ns_jft_in1k
A EfficientNet image classification model. Trained on ImageNet-1k and unlabeled JFT-300m using Noisy Student semi-supervised learning in Tensorflow by paper authors, ported to PyTorch by Ross Wightman.
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 30.4
- GMACs: 10.5
- Activations (M): 98.9
- Image size: 456 x 456
- **Papers:**
- EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks: https://arxiv.org/abs/1905.11946
- Self-training with Noisy Student improves ImageNet classification: https://arxiv.org/abs/1911.04252
- **Dataset:** ImageNet-1k
- **Original:** https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('tf_efficientnet_b5.ns_jft_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Feature Map Extraction
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'tf_efficientnet_b5.ns_jft_in1k',
pretrained=True,
features_only=True,
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
for o in output:
# print shape of each feature map in output
# e.g.:
# torch.Size([1, 24, 228, 228])
# torch.Size([1, 40, 114, 114])
# torch.Size([1, 64, 57, 57])
# torch.Size([1, 176, 29, 29])
# torch.Size([1, 512, 15, 15])
print(o.shape)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'tf_efficientnet_b5.ns_jft_in1k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 2048, 15, 15) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
## Citation
```bibtex
@inproceedings{tan2019efficientnet,
title={Efficientnet: Rethinking model scaling for convolutional neural networks},
author={Tan, Mingxing and Le, Quoc},
booktitle={International conference on machine learning},
pages={6105--6114},
year={2019},
organization={PMLR}
}
```
```bibtex
@article{Xie2019SelfTrainingWN,
title={Self-Training With Noisy Student Improves ImageNet Classification},
author={Qizhe Xie and Eduard H. Hovy and Minh-Thang Luong and Quoc V. Le},
journal={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
year={2019},
pages={10684-10695}
}
```
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}
```
| 4,606 | [
[
-0.0298919677734375,
-0.04205322265625,
-0.006832122802734375,
0.00911712646484375,
-0.01806640625,
-0.0279998779296875,
-0.02520751953125,
-0.031494140625,
0.0115203857421875,
0.026397705078125,
-0.0265045166015625,
-0.042022705078125,
-0.054901123046875,
-0.01030731201171875,
-0.0197906494140625,
0.06591796875,
-0.0056915283203125,
-9.5367431640625e-7,
-0.01219940185546875,
-0.041259765625,
-0.00302886962890625,
-0.017059326171875,
-0.07037353515625,
-0.033050537109375,
0.02520751953125,
0.018798828125,
0.0401611328125,
0.050811767578125,
0.04962158203125,
0.039154052734375,
-0.00782012939453125,
0.0036869049072265625,
-0.021148681640625,
-0.00826263427734375,
0.0311737060546875,
-0.04071044921875,
-0.027130126953125,
0.012908935546875,
0.05621337890625,
0.038604736328125,
-0.0029735565185546875,
0.036773681640625,
0.0116424560546875,
0.044342041015625,
-0.023956298828125,
0.01499176025390625,
-0.0277099609375,
0.01412200927734375,
-0.005695343017578125,
0.0115814208984375,
-0.0225067138671875,
-0.022674560546875,
0.017608642578125,
-0.04132080078125,
0.03875732421875,
-0.006011962890625,
0.097900390625,
0.022735595703125,
-0.006023406982421875,
0.001194000244140625,
-0.016021728515625,
0.056182861328125,
-0.062286376953125,
0.0184173583984375,
0.01045989990234375,
0.014862060546875,
-0.00728607177734375,
-0.08306884765625,
-0.035797119140625,
-0.014373779296875,
-0.017333984375,
-0.0076904296875,
-0.0256195068359375,
0.01123809814453125,
0.0284881591796875,
0.022552490234375,
-0.0304718017578125,
0.006824493408203125,
-0.042755126953125,
-0.0159454345703125,
0.044189453125,
-0.001949310302734375,
0.016510009765625,
-0.01123809814453125,
-0.033203125,
-0.035064697265625,
-0.0181732177734375,
0.029388427734375,
0.0187225341796875,
0.020538330078125,
-0.03912353515625,
0.02197265625,
0.004299163818359375,
0.0513916015625,
0.014068603515625,
-0.0291900634765625,
0.048187255859375,
0.00321197509765625,
-0.035736083984375,
-0.0101776123046875,
0.08551025390625,
0.02581787109375,
0.0183563232421875,
0.00389862060546875,
-0.01245880126953125,
-0.03497314453125,
-0.0013971328735351562,
-0.096923828125,
-0.028961181640625,
0.0234527587890625,
-0.052154541015625,
-0.033843994140625,
0.01171875,
-0.041473388671875,
-0.006488800048828125,
-0.006145477294921875,
0.05181884765625,
-0.028350830078125,
-0.03594970703125,
-0.0013065338134765625,
-0.01210784912109375,
0.01070404052734375,
0.0220947265625,
-0.041168212890625,
0.0118255615234375,
0.0171356201171875,
0.084228515625,
0.0057373046875,
-0.033538818359375,
-0.01490020751953125,
-0.033721923828125,
-0.0202484130859375,
0.02777099609375,
-0.0008616447448730469,
-0.0015573501586914062,
-0.024658203125,
0.0253753662109375,
-0.01126861572265625,
-0.054443359375,
0.022918701171875,
-0.0165557861328125,
0.012054443359375,
0.006275177001953125,
-0.0220489501953125,
-0.04132080078125,
0.0211029052734375,
-0.035614013671875,
0.08734130859375,
0.0268402099609375,
-0.06378173828125,
0.0210418701171875,
-0.04071044921875,
-0.0113983154296875,
-0.0192718505859375,
0.004177093505859375,
-0.0865478515625,
-0.0070648193359375,
0.01445770263671875,
0.0682373046875,
-0.0178375244140625,
0.0110015869140625,
-0.0474853515625,
-0.0190277099609375,
0.02294921875,
-0.00846099853515625,
0.081298828125,
0.0214996337890625,
-0.035552978515625,
0.0207977294921875,
-0.048187255859375,
0.016265869140625,
0.03643798828125,
-0.0188446044921875,
-0.0025463104248046875,
-0.047515869140625,
0.0113983154296875,
0.0205230712890625,
0.0120086669921875,
-0.03912353515625,
0.01447296142578125,
-0.011077880859375,
0.03753662109375,
0.045989990234375,
-0.0104217529296875,
0.0310516357421875,
-0.0254669189453125,
0.0179901123046875,
0.022125244140625,
0.0185546875,
-0.0045166015625,
-0.030731201171875,
-0.06317138671875,
-0.0391845703125,
0.026153564453125,
0.018768310546875,
-0.0364990234375,
0.030303955078125,
-0.01552581787109375,
-0.060333251953125,
-0.033905029296875,
0.0098419189453125,
0.032928466796875,
0.052886962890625,
0.02679443359375,
-0.026580810546875,
-0.032440185546875,
-0.0689697265625,
-0.0007495880126953125,
-0.0011444091796875,
0.0030460357666015625,
0.023284912109375,
0.04510498046875,
-0.001453399658203125,
0.040557861328125,
-0.0247650146484375,
-0.0240478515625,
-0.0166015625,
0.00872039794921875,
0.034912109375,
0.0633544921875,
0.05865478515625,
-0.046630859375,
-0.043792724609375,
-0.01532745361328125,
-0.0711669921875,
0.007328033447265625,
-0.0112762451171875,
-0.0120391845703125,
0.01416015625,
0.02105712890625,
-0.03948974609375,
0.03680419921875,
0.017974853515625,
-0.02740478515625,
0.02728271484375,
-0.0166015625,
0.01500701904296875,
-0.081298828125,
0.0080108642578125,
0.033416748046875,
-0.0167694091796875,
-0.0406494140625,
0.00974273681640625,
0.007419586181640625,
-0.0011005401611328125,
-0.0347900390625,
0.044097900390625,
-0.043212890625,
-0.007843017578125,
-0.0104217529296875,
-0.0261077880859375,
-0.00021004676818847656,
0.057464599609375,
-0.0095367431640625,
0.0307769775390625,
0.063232421875,
-0.03497314453125,
0.031494140625,
0.0185089111328125,
-0.0148162841796875,
0.0273284912109375,
-0.0555419921875,
0.00891876220703125,
0.0025920867919921875,
0.0199432373046875,
-0.07537841796875,
-0.0161895751953125,
0.0240478515625,
-0.043975830078125,
0.05126953125,
-0.040191650390625,
-0.031463623046875,
-0.03863525390625,
-0.029144287109375,
0.0289459228515625,
0.0467529296875,
-0.060546875,
0.03363037109375,
0.0190887451171875,
0.029815673828125,
-0.0447998046875,
-0.06695556640625,
-0.0197601318359375,
-0.031585693359375,
-0.058074951171875,
0.0234375,
0.010406494140625,
0.00997161865234375,
0.008819580078125,
-0.0020236968994140625,
-0.01116943359375,
0.00342559814453125,
0.037933349609375,
0.0207672119140625,
-0.0212860107421875,
0.0014400482177734375,
-0.02044677734375,
0.004138946533203125,
0.0081634521484375,
-0.02777099609375,
0.03460693359375,
-0.0258636474609375,
-0.0005583763122558594,
-0.06195068359375,
-0.00498199462890625,
0.035247802734375,
-0.00029850006103515625,
0.0615234375,
0.08984375,
-0.034393310546875,
-0.008087158203125,
-0.0309295654296875,
-0.02227783203125,
-0.03839111328125,
0.039276123046875,
-0.0261993408203125,
-0.04583740234375,
0.0594482421875,
-0.0045166015625,
0.0091094970703125,
0.055694580078125,
0.0262451171875,
-0.00795745849609375,
0.047576904296875,
0.0428466796875,
0.017120361328125,
0.06158447265625,
-0.079345703125,
-0.015106201171875,
-0.0594482421875,
-0.027618408203125,
-0.0278472900390625,
-0.053680419921875,
-0.0565185546875,
-0.0218963623046875,
0.037200927734375,
0.017242431640625,
-0.043609619140625,
0.031280517578125,
-0.06805419921875,
0.0066986083984375,
0.047821044921875,
0.043548583984375,
-0.026031494140625,
0.02490234375,
-0.0117645263671875,
0.0031070709228515625,
-0.0625,
-0.009979248046875,
0.08935546875,
0.0355224609375,
0.048187255859375,
-0.01032257080078125,
0.05535888671875,
-0.0161590576171875,
0.0265350341796875,
-0.04583740234375,
0.043426513671875,
-0.01137542724609375,
-0.033294677734375,
-0.0200042724609375,
-0.0447998046875,
-0.081298828125,
0.014862060546875,
-0.0218658447265625,
-0.0565185546875,
0.01763916015625,
0.01580810546875,
-0.02008056640625,
0.058319091796875,
-0.0704345703125,
0.07269287109375,
-0.00543975830078125,
-0.03814697265625,
0.004058837890625,
-0.0504150390625,
0.0218963623046875,
0.015777587890625,
-0.0195159912109375,
-0.00505828857421875,
0.007297515869140625,
0.08721923828125,
-0.049957275390625,
0.063232421875,
-0.043426513671875,
0.033447265625,
0.042266845703125,
-0.0081329345703125,
0.0277099609375,
-0.0083160400390625,
-0.012481689453125,
0.032958984375,
0.0014181137084960938,
-0.036529541015625,
-0.040313720703125,
0.045013427734375,
-0.07952880859375,
-0.0259552001953125,
-0.0201568603515625,
-0.03741455078125,
0.0164642333984375,
0.01117706298828125,
0.03643798828125,
0.048095703125,
0.02154541015625,
0.0275726318359375,
0.041473388671875,
-0.0198974609375,
0.042388916015625,
-0.005908966064453125,
-0.01007843017578125,
-0.033355712890625,
0.061004638671875,
0.0272674560546875,
0.0145111083984375,
0.00811004638671875,
0.0209197998046875,
-0.022308349609375,
-0.04315185546875,
-0.02532958984375,
0.01947021484375,
-0.053314208984375,
-0.042205810546875,
-0.05303955078125,
-0.03485107421875,
-0.0279388427734375,
-0.00801849365234375,
-0.0419921875,
-0.03350830078125,
-0.03375244140625,
0.0158843994140625,
0.052886962890625,
0.037109375,
-0.0153656005859375,
0.04547119140625,
-0.033538818359375,
0.0045623779296875,
0.01036834716796875,
0.0316162109375,
0.008514404296875,
-0.07012939453125,
-0.02447509765625,
-0.0093841552734375,
-0.03515625,
-0.045501708984375,
0.03851318359375,
0.0203704833984375,
0.03826904296875,
0.0305023193359375,
-0.00940704345703125,
0.05523681640625,
0.00547027587890625,
0.037109375,
0.03277587890625,
-0.042449951171875,
0.037567138671875,
-0.00213623046875,
0.0171356201171875,
0.01250457763671875,
0.0225982666015625,
-0.011688232421875,
-0.007049560546875,
-0.0810546875,
-0.055694580078125,
0.0650634765625,
0.006504058837890625,
0.0010852813720703125,
0.0323486328125,
0.055145263671875,
-0.0004391670227050781,
0.0021610260009765625,
-0.05780029296875,
-0.0355224609375,
-0.028839111328125,
-0.023223876953125,
0.0007915496826171875,
0.0002892017364501953,
-0.000583648681640625,
-0.055023193359375,
0.04693603515625,
-0.0093231201171875,
0.061981201171875,
0.02862548828125,
-0.0024700164794921875,
-0.01091766357421875,
-0.0289154052734375,
0.0279083251953125,
0.0194091796875,
-0.02532958984375,
0.010650634765625,
0.0162811279296875,
-0.0421142578125,
0.0123443603515625,
0.01422882080078125,
-0.003482818603515625,
0.00033473968505859375,
0.040557861328125,
0.069091796875,
-0.004390716552734375,
0.010284423828125,
0.03472900390625,
-0.00629425048828125,
-0.03662109375,
-0.0185546875,
0.017059326171875,
-0.0052490234375,
0.0382080078125,
0.0242919921875,
0.03363037109375,
-0.0037975311279296875,
-0.01349639892578125,
0.019195556640625,
0.038421630859375,
-0.019561767578125,
-0.0213775634765625,
0.04998779296875,
-0.012451171875,
-0.0085601806640625,
0.0655517578125,
-0.01229095458984375,
-0.037261962890625,
0.084228515625,
0.031402587890625,
0.07208251953125,
-0.0003273487091064453,
0.0010814666748046875,
0.0775146484375,
0.016510009765625,
-0.006389617919921875,
0.00672149658203125,
0.0025577545166015625,
-0.057098388671875,
0.0030422210693359375,
-0.041839599609375,
0.00359344482421875,
0.02362060546875,
-0.036529541015625,
0.0189056396484375,
-0.05621337890625,
-0.03216552734375,
0.015350341796875,
0.03021240234375,
-0.0709228515625,
0.0169525146484375,
-0.003910064697265625,
0.06207275390625,
-0.054412841796875,
0.06072998046875,
0.06121826171875,
-0.034576416015625,
-0.09149169921875,
-0.01059722900390625,
-0.005863189697265625,
-0.061004638671875,
0.04461669921875,
0.0355224609375,
0.013702392578125,
0.00830078125,
-0.0665283203125,
-0.05340576171875,
0.10797119140625,
0.04071044921875,
-0.01297760009765625,
0.0207366943359375,
-0.00971221923828125,
0.02105712890625,
-0.033355712890625,
0.04229736328125,
0.0125274658203125,
0.0270233154296875,
0.0226287841796875,
-0.05035400390625,
0.022735595703125,
-0.026153564453125,
0.004322052001953125,
0.0146942138671875,
-0.072265625,
0.07122802734375,
-0.0382080078125,
-0.00672149658203125,
-0.005344390869140625,
0.0577392578125,
0.007904052734375,
0.01190948486328125,
0.04931640625,
0.05975341796875,
0.04425048828125,
-0.023651123046875,
0.0633544921875,
0.0028438568115234375,
0.0516357421875,
0.046630859375,
0.0423583984375,
0.035797119140625,
0.027862548828125,
-0.022796630859375,
0.0198974609375,
0.0814208984375,
-0.02960205078125,
0.021575927734375,
0.016845703125,
0.006511688232421875,
-0.01548004150390625,
0.007747650146484375,
-0.0258331298828125,
0.04071044921875,
0.01031494140625,
-0.041900634765625,
-0.019317626953125,
0.00439453125,
0.0010557174682617188,
-0.0300750732421875,
-0.022491455078125,
0.033294677734375,
0.002132415771484375,
-0.0287322998046875,
0.0714111328125,
0.003360748291015625,
0.06951904296875,
-0.0260467529296875,
0.006031036376953125,
-0.020355224609375,
0.0193023681640625,
-0.0294342041015625,
-0.05877685546875,
0.023406982421875,
-0.02105712890625,
0.002422332763671875,
-0.0013484954833984375,
0.054595947265625,
-0.029296875,
-0.038604736328125,
0.016326904296875,
0.0231170654296875,
0.036865234375,
0.0017910003662109375,
-0.0966796875,
0.0100250244140625,
0.00443267822265625,
-0.05731201171875,
0.02423095703125,
0.033355712890625,
0.007843017578125,
0.05731201171875,
0.04168701171875,
-0.00998687744140625,
0.011322021484375,
-0.0110931396484375,
0.0614013671875,
-0.029388427734375,
-0.019317626953125,
-0.05865478515625,
0.04498291015625,
-0.00843048095703125,
-0.04437255859375,
0.032257080078125,
0.03460693359375,
0.0673828125,
0.002544403076171875,
0.0258636474609375,
-0.0219879150390625,
-0.003810882568359375,
-0.0213470458984375,
0.05975341796875,
-0.062255859375,
-0.0035076141357421875,
-0.01326751708984375,
-0.046661376953125,
-0.02923583984375,
0.055938720703125,
-0.014129638671875,
0.039031982421875,
0.032440185546875,
0.07537841796875,
-0.0258636474609375,
-0.026397705078125,
0.0200958251953125,
0.01513671875,
0.00905609130859375,
0.032928466796875,
0.025299072265625,
-0.0614013671875,
0.031524658203125,
-0.05877685546875,
-0.01494598388671875,
-0.01202392578125,
-0.0478515625,
-0.0706787109375,
-0.0679931640625,
-0.05133056640625,
-0.050384521484375,
-0.0168914794921875,
0.07379150390625,
0.0838623046875,
-0.049835205078125,
-0.01268768310546875,
-0.00299072265625,
0.01329803466796875,
-0.0234527587890625,
-0.0178680419921875,
0.051666259765625,
-0.019378662109375,
-0.055419921875,
-0.029876708984375,
-0.007617950439453125,
0.0268402099609375,
-0.002288818359375,
-0.015106201171875,
-0.0129547119140625,
-0.026885986328125,
0.01262664794921875,
0.0172882080078125,
-0.04229736328125,
-0.00997161865234375,
-0.0194549560546875,
-0.014404296875,
0.028900146484375,
0.027862548828125,
-0.039215087890625,
0.028167724609375,
0.02972412109375,
0.0291595458984375,
0.06488037109375,
-0.02972412109375,
-0.001979827880859375,
-0.0587158203125,
0.041229248046875,
-0.0112457275390625,
0.034271240234375,
0.03167724609375,
-0.033843994140625,
0.049163818359375,
0.02923583984375,
-0.03826904296875,
-0.0614013671875,
-0.0202178955078125,
-0.08154296875,
-0.01088714599609375,
0.06915283203125,
-0.0384521484375,
-0.04168701171875,
0.037567138671875,
0.004367828369140625,
0.053070068359375,
-0.0164642333984375,
0.038055419921875,
0.0156707763671875,
-0.01006317138671875,
-0.053009033203125,
-0.038543701171875,
0.02972412109375,
0.0149383544921875,
-0.039886474609375,
-0.030609130859375,
-0.0032901763916015625,
0.05108642578125,
0.0146484375,
0.03497314453125,
-0.00244903564453125,
0.0107421875,
0.00824737548828125,
0.036590576171875,
-0.0396728515625,
-0.0034732818603515625,
-0.0299224853515625,
0.01186370849609375,
-0.00617218017578125,
-0.0419921875
]
] |
zarakiquemparte/kuchiki-l2-7b | 2023-09-14T13:48:00.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama2",
"license:other",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | zarakiquemparte | null | null | zarakiquemparte/kuchiki-l2-7b | 7 | 7,415 | transformers | 2023-08-12T15:37:50 | ---
license: other
tags:
- llama2
---
# Model Card: Kuchiki L2 7b
This model uses [Nous Hermes Llama2 7b](https://huggingface.co/NousResearch/Nous-Hermes-llama-2-7b) (70%) as a base with [Airoboros L2 7B GPT4 2.0](https://huggingface.co/jondurbin/airoboros-l2-7b-gpt4-2.0) (30%) and the result of this merge was merged with [LimaRP LLama2 7B Lora](https://huggingface.co/lemonilia/limarp-llama2).
This merge of models(hermes and airoboros) was done with this [script](https://github.com/zarakiquemparte/zaraki-tools/blob/main/merge-cli.py)
This merge of Lora with Model was done with this [script](https://github.com/zarakiquemparte/zaraki-tools/blob/main/apply-lora.py)
Quantized Model by @TheBloke:
- [GGUF](https://huggingface.co/TheBloke/Kuchiki-L2-7B-GGUF)
- [GPTQ](https://huggingface.co/TheBloke/Kuchiki-L2-7B-GPTQ)
Merge illustration:
## Usage:
Since this is a merge between Nous Hermes, Airoboros and LimaRP, the following instruction formats should work:
Alpaca 2:
```
### Instruction:
<prompt>
### Response:
<leave a newline blank for model to respond>
```
LimaRP instruction format:
```
<<SYSTEM>>
<character card and system prompt>
<<USER>>
<prompt>
<<AIBOT>>
<leave a newline blank for model to respond>
```
## Bias, Risks, and Limitations
This model is not intended for supplying factual information or advice in any form
## Training Details
This model is merged and can be reproduced using the tools mentioned above. Please refer to all provided links for extra model-specific details. | 1,559 | [
[
-0.029510498046875,
-0.03790283203125,
0.0311126708984375,
0.0226898193359375,
-0.042999267578125,
-0.0200347900390625,
0.0182342529296875,
-0.04193115234375,
0.0253753662109375,
0.060546875,
-0.052825927734375,
-0.037445068359375,
-0.043701171875,
-0.01271820068359375,
-0.028533935546875,
0.0966796875,
-0.0012826919555664062,
-0.0135498046875,
0.003757476806640625,
-0.0207977294921875,
-0.0159759521484375,
-0.03729248046875,
-0.0682373046875,
-0.034912109375,
0.0301971435546875,
0.0161285400390625,
0.060272216796875,
0.0243682861328125,
0.03875732421875,
0.0267181396484375,
-0.0283660888671875,
0.01300811767578125,
-0.0178070068359375,
-0.014617919921875,
0.0130462646484375,
-0.0341796875,
-0.087158203125,
0.0008382797241210938,
0.04168701171875,
0.0302886962890625,
-0.034149169921875,
0.01300811767578125,
-0.00848388671875,
0.0281829833984375,
-0.033599853515625,
0.002593994140625,
-0.0189361572265625,
0.00881195068359375,
0.009613037109375,
0.0049896240234375,
-0.01270294189453125,
-0.038818359375,
-0.0005288124084472656,
-0.06964111328125,
0.007061004638671875,
-0.01708984375,
0.085205078125,
0.014373779296875,
-0.049407958984375,
-0.0157623291015625,
-0.034576416015625,
0.050506591796875,
-0.07855224609375,
0.008758544921875,
0.02685546875,
0.0191802978515625,
-0.0216827392578125,
-0.049652099609375,
-0.04888916015625,
0.0025634765625,
-0.0015697479248046875,
0.011566162109375,
-0.03509521484375,
-0.00954437255859375,
0.03729248046875,
0.0303497314453125,
-0.036346435546875,
0.0115509033203125,
-0.0653076171875,
-0.00823974609375,
0.048065185546875,
0.02728271484375,
0.006305694580078125,
-0.01259613037109375,
-0.041046142578125,
-0.03173828125,
-0.036376953125,
0.00015926361083984375,
0.042724609375,
0.00775909423828125,
-0.0626220703125,
0.0648193359375,
-0.0164794921875,
0.0557861328125,
0.0207977294921875,
-0.0281524658203125,
0.0382080078125,
-0.01125335693359375,
-0.037078857421875,
-0.00400543212890625,
0.067138671875,
0.0374755859375,
-0.009796142578125,
0.0216827392578125,
-0.007755279541015625,
0.0017337799072265625,
-0.00008171796798706055,
-0.0662841796875,
0.00567626953125,
0.02923583984375,
-0.030303955078125,
-0.043609619140625,
0.0004401206970214844,
-0.041046142578125,
-0.0085296630859375,
0.0008916854858398438,
0.03302001953125,
-0.02496337890625,
-0.024078369140625,
0.01280975341796875,
-0.01103973388671875,
0.048919677734375,
0.02984619140625,
-0.057586669921875,
0.0201416015625,
0.037872314453125,
0.044464111328125,
0.028076171875,
-0.0102996826171875,
-0.024871826171875,
0.01560211181640625,
-0.02984619140625,
0.050018310546875,
-0.00493621826171875,
-0.04693603515625,
-0.01120758056640625,
0.0105438232421875,
0.0088348388671875,
-0.035614013671875,
0.050018310546875,
-0.038330078125,
0.030303955078125,
-0.01221466064453125,
-0.02142333984375,
-0.031341552734375,
0.0207977294921875,
-0.064208984375,
0.07940673828125,
0.0276031494140625,
-0.057586669921875,
-0.001461029052734375,
-0.049346923828125,
-0.01434326171875,
-0.0140533447265625,
0.0064697265625,
-0.058074951171875,
-0.005680084228515625,
0.00516510009765625,
0.0244903564453125,
-0.0272064208984375,
0.00047969818115234375,
-0.04248046875,
-0.0241241455078125,
0.01873779296875,
-0.0021762847900390625,
0.0697021484375,
0.020965576171875,
-0.0092010498046875,
0.00376129150390625,
-0.0670166015625,
0.0149993896484375,
0.0241546630859375,
-0.031402587890625,
-0.00939178466796875,
-0.039306640625,
0.0121002197265625,
0.00496673583984375,
0.032073974609375,
-0.01253509521484375,
0.05426025390625,
-0.016204833984375,
0.02789306640625,
0.0399169921875,
0.00539398193359375,
0.027008056640625,
-0.036102294921875,
0.0321044921875,
0.0159149169921875,
0.0245819091796875,
0.00021767616271972656,
-0.0634765625,
-0.07122802734375,
-0.0106964111328125,
0.00711822509765625,
0.0301971435546875,
-0.03411865234375,
0.043701171875,
-0.0030040740966796875,
-0.053924560546875,
-0.031768798828125,
-0.01184844970703125,
0.01239776611328125,
0.038665771484375,
0.019500732421875,
-0.0239410400390625,
-0.046356201171875,
-0.06988525390625,
0.0231170654296875,
-0.0193023681640625,
-0.0025081634521484375,
0.002574920654296875,
0.043853759765625,
-0.03656005859375,
0.03802490234375,
-0.043609619140625,
-0.01320648193359375,
-0.01788330078125,
0.0212554931640625,
0.046905517578125,
0.052032470703125,
0.047637939453125,
-0.04254150390625,
-0.02655029296875,
0.0002751350402832031,
-0.0576171875,
-0.0006656646728515625,
0.0169830322265625,
-0.01678466796875,
0.0215301513671875,
0.0034427642822265625,
-0.07080078125,
0.051116943359375,
0.046630859375,
-0.03302001953125,
0.022613525390625,
-0.025909423828125,
0.020355224609375,
-0.1058349609375,
0.027862548828125,
-0.007568359375,
0.0018749237060546875,
-0.050933837890625,
0.02191162109375,
0.00926971435546875,
-0.0126190185546875,
-0.042572021484375,
0.05999755859375,
-0.030731201171875,
-0.009613037109375,
-0.0246734619140625,
-0.010101318359375,
0.01412200927734375,
0.033935546875,
-0.0004584789276123047,
0.0218505859375,
0.033782958984375,
-0.042572021484375,
0.042572021484375,
0.036773681640625,
-0.008392333984375,
0.037445068359375,
-0.053375244140625,
0.01788330078125,
0.0023040771484375,
0.031890869140625,
-0.0625,
-0.027191162109375,
0.0535888671875,
-0.0232391357421875,
0.0067596435546875,
-0.016845703125,
-0.03271484375,
-0.03009033203125,
-0.0272369384765625,
0.03875732421875,
0.05877685546875,
-0.036712646484375,
0.06683349609375,
-0.00409698486328125,
0.009063720703125,
-0.05316162109375,
-0.054168701171875,
-0.0301666259765625,
-0.0307159423828125,
-0.049102783203125,
0.0247955322265625,
-0.0165557861328125,
-0.0037021636962890625,
0.017578125,
-0.005519866943359375,
-0.01983642578125,
-0.01026153564453125,
0.03314208984375,
0.055450439453125,
-0.0198822021484375,
-0.02899169921875,
0.016693115234375,
-0.007747650146484375,
-0.016998291015625,
0.01290130615234375,
0.05035400390625,
-0.00250244140625,
-0.020904541015625,
-0.0469970703125,
0.0165252685546875,
0.06634521484375,
-0.0092315673828125,
0.0684814453125,
0.053955078125,
-0.02813720703125,
0.0122833251953125,
-0.05364990234375,
0.0018558502197265625,
-0.029266357421875,
0.01251983642578125,
-0.0297393798828125,
-0.03839111328125,
0.06671142578125,
0.02276611328125,
-0.00513458251953125,
0.048919677734375,
0.033843994140625,
0.0034961700439453125,
0.068115234375,
0.057037353515625,
-0.014892578125,
0.02557373046875,
-0.058990478515625,
0.01288604736328125,
-0.08251953125,
-0.036346435546875,
-0.025848388671875,
-0.01003265380859375,
-0.046875,
-0.03594970703125,
0.02020263671875,
0.0330810546875,
-0.028656005859375,
0.045623779296875,
-0.0269622802734375,
0.0121002197265625,
0.030517578125,
0.01152801513671875,
0.029541015625,
0.0171356201171875,
0.00917816162109375,
0.0254364013671875,
-0.04083251953125,
-0.02191162109375,
0.09478759765625,
0.02679443359375,
0.060394287109375,
0.018768310546875,
0.0625,
-0.00702667236328125,
0.021148681640625,
-0.0411376953125,
0.0458984375,
0.0020160675048828125,
-0.03729248046875,
-0.00676727294921875,
-0.02935791015625,
-0.06915283203125,
0.0237884521484375,
-0.01113128662109375,
-0.041259765625,
0.015869140625,
0.0005693435668945312,
-0.052642822265625,
0.015045166015625,
-0.039031982421875,
0.05047607421875,
-0.00237274169921875,
-0.01183319091796875,
-0.0177154541015625,
-0.031768798828125,
0.051910400390625,
-0.0015583038330078125,
0.002407073974609375,
-0.0117950439453125,
-0.0126800537109375,
0.06500244140625,
-0.0478515625,
0.050018310546875,
0.004924774169921875,
-0.030029296875,
0.03521728515625,
-0.0233154296875,
0.040863037109375,
-0.0038890838623046875,
0.0006814002990722656,
0.0179443359375,
0.0030918121337890625,
-0.0267333984375,
-0.0462646484375,
0.058746337890625,
-0.08038330078125,
-0.0209808349609375,
-0.03778076171875,
-0.025238037109375,
0.01233673095703125,
0.0007815361022949219,
0.036865234375,
0.0291595458984375,
0.005008697509765625,
-0.007534027099609375,
0.036529541015625,
-0.01222991943359375,
0.0223388671875,
0.06317138671875,
-0.025726318359375,
-0.0555419921875,
0.035247802734375,
-0.01548004150390625,
0.0212249755859375,
0.00865936279296875,
-0.0006194114685058594,
-0.0192718505859375,
0.0014123916625976562,
-0.0291290283203125,
0.051300048828125,
-0.05902099609375,
-0.0257720947265625,
-0.0267486572265625,
-0.03802490234375,
-0.04656982421875,
0.0024204254150390625,
-0.023162841796875,
-0.037322998046875,
-0.02679443359375,
0.00043320655822753906,
0.03680419921875,
0.051544189453125,
-0.024383544921875,
0.05072021484375,
-0.06329345703125,
0.0279998779296875,
0.02587890625,
0.006023406982421875,
0.00658416748046875,
-0.05999755859375,
-0.0011577606201171875,
0.00942230224609375,
-0.035675048828125,
-0.08447265625,
0.042449951171875,
-0.0108642578125,
0.03924560546875,
0.0298309326171875,
-0.009613037109375,
0.06304931640625,
0.002750396728515625,
0.042816162109375,
0.0258331298828125,
-0.06109619140625,
0.048248291015625,
-0.0289306640625,
0.01177978515625,
0.001384735107421875,
0.038787841796875,
-0.0384521484375,
-0.01140594482421875,
-0.071533203125,
-0.052398681640625,
0.06878662109375,
0.024139404296875,
-0.00885772705078125,
0.0200347900390625,
0.02203369140625,
-0.019378662109375,
0.0222625732421875,
-0.069580078125,
-0.031402587890625,
-0.0105743408203125,
-0.00787353515625,
0.0026073455810546875,
-0.0202178955078125,
-0.032806396484375,
-0.01244354248046875,
0.066162109375,
-0.005680084228515625,
0.02587890625,
0.005939483642578125,
0.0250091552734375,
-0.0289154052734375,
0.007221221923828125,
0.05419921875,
0.0096893310546875,
-0.03302001953125,
-0.0216064453125,
0.01161956787109375,
-0.0211181640625,
0.0082550048828125,
0.01519012451171875,
-0.002315521240234375,
-0.01187896728515625,
0.03887939453125,
0.07159423828125,
0.01385498046875,
-0.024566650390625,
0.029266357421875,
-0.007228851318359375,
-0.027191162109375,
-0.01508331298828125,
0.00942230224609375,
0.0159759521484375,
0.02996826171875,
0.00926971435546875,
0.008270263671875,
-0.003452301025390625,
-0.05377197265625,
-0.018218994140625,
0.03240966796875,
0.0035228729248046875,
-0.0159149169921875,
0.045379638671875,
0.00806427001953125,
-0.004314422607421875,
0.05908203125,
-0.00884246826171875,
-0.0282135009765625,
0.0682373046875,
0.037567138671875,
0.0562744140625,
-0.0266265869140625,
0.00006312131881713867,
0.05108642578125,
0.005828857421875,
-0.017333984375,
0.029083251953125,
-0.00782012939453125,
-0.0498046875,
-0.0174713134765625,
-0.039947509765625,
-0.029571533203125,
0.04449462890625,
-0.056793212890625,
0.039337158203125,
-0.03369140625,
-0.0236663818359375,
-0.0153045654296875,
0.0142669677734375,
-0.0404052734375,
0.01544952392578125,
0.01399993896484375,
0.05487060546875,
-0.0855712890625,
0.075927734375,
0.0509033203125,
-0.047393798828125,
-0.09173583984375,
-0.0256500244140625,
-0.0098114013671875,
-0.0758056640625,
0.05218505859375,
-0.0007719993591308594,
0.0190582275390625,
-0.0222930908203125,
-0.041900634765625,
-0.08251953125,
0.10272216796875,
0.023712158203125,
-0.02850341796875,
0.0007982254028320312,
0.0016117095947265625,
0.03033447265625,
-0.0232696533203125,
0.041290283203125,
0.0296478271484375,
0.028106689453125,
0.042724609375,
-0.08380126953125,
0.0038356781005859375,
-0.0226898193359375,
-0.0028476715087890625,
0.0106964111328125,
-0.07098388671875,
0.09124755859375,
-0.0245819091796875,
0.00029158592224121094,
0.0557861328125,
0.046539306640625,
0.0511474609375,
0.008453369140625,
0.0249786376953125,
0.0806884765625,
0.04278564453125,
-0.0008683204650878906,
0.0540771484375,
-0.01763916015625,
0.04925537109375,
0.0692138671875,
-0.027587890625,
0.07080078125,
0.027069091796875,
-0.012054443359375,
0.05303955078125,
0.045013427734375,
-0.0183563232421875,
0.037017822265625,
-0.004364013671875,
-0.0193023681640625,
-0.006534576416015625,
0.0126495361328125,
-0.07135009765625,
0.031768798828125,
0.01491546630859375,
-0.0223388671875,
-0.0194854736328125,
-0.022125244140625,
-0.00203704833984375,
-0.034393310546875,
-0.01422119140625,
0.039520263671875,
0.0002884864807128906,
-0.052398681640625,
0.0550537109375,
0.017730712890625,
0.047943115234375,
-0.06927490234375,
-0.0253143310546875,
-0.0404052734375,
0.015899658203125,
-0.0284271240234375,
-0.041473388671875,
0.0128936767578125,
-0.00736236572265625,
-0.012176513671875,
0.0180511474609375,
0.039886474609375,
-0.034820556640625,
-0.049468994140625,
0.0270538330078125,
0.010986328125,
0.0221405029296875,
0.01358795166015625,
-0.057281494140625,
0.035369873046875,
0.00510406494140625,
-0.0158843994140625,
0.03253173828125,
0.0191192626953125,
0.0012187957763671875,
0.06005859375,
0.052154541015625,
-0.00133514404296875,
-0.0043792724609375,
-0.00757598876953125,
0.0706787109375,
-0.036651611328125,
-0.03802490234375,
-0.04644775390625,
0.045166015625,
-0.0001970529556274414,
-0.0264129638671875,
0.036224365234375,
0.055938720703125,
0.040130615234375,
-0.0175018310546875,
0.05084228515625,
-0.003726959228515625,
0.0194549560546875,
-0.04583740234375,
0.05242919921875,
-0.052459716796875,
0.01213836669921875,
-0.012451171875,
-0.076904296875,
0.0066070556640625,
0.0594482421875,
0.0029201507568359375,
0.00554656982421875,
0.04022216796875,
0.0650634765625,
-0.00861358642578125,
-0.01438140869140625,
0.019744873046875,
0.01800537109375,
0.01436614990234375,
0.061126708984375,
0.061798095703125,
-0.0816650390625,
0.03302001953125,
-0.03228759765625,
-0.01233673095703125,
-0.0192718505859375,
-0.06591796875,
-0.05938720703125,
-0.0234527587890625,
-0.0352783203125,
-0.03265380859375,
-0.018218994140625,
0.058135986328125,
0.0595703125,
-0.04254150390625,
-0.034271240234375,
0.01238250732421875,
-0.0025997161865234375,
-0.00011348724365234375,
-0.01322174072265625,
0.02117919921875,
0.0157318115234375,
-0.062103271484375,
0.0129241943359375,
0.00043654441833496094,
0.051055908203125,
-0.003635406494140625,
-0.022705078125,
0.005489349365234375,
-0.00121307373046875,
0.0251312255859375,
0.032440185546875,
-0.07269287109375,
0.009307861328125,
-0.0174560546875,
-0.01012420654296875,
-0.0054931640625,
0.0292816162109375,
-0.046356201171875,
-0.00821685791015625,
0.0248260498046875,
0.00122833251953125,
0.04339599609375,
-0.0018053054809570312,
0.02203369140625,
-0.031707763671875,
0.0302581787109375,
-0.019195556640625,
0.038116455078125,
0.0203704833984375,
-0.0230255126953125,
0.0482177734375,
0.01849365234375,
-0.029388427734375,
-0.06329345703125,
-0.00616455078125,
-0.1138916015625,
-0.004062652587890625,
0.07196044921875,
0.0014524459838867188,
-0.02294921875,
0.045318603515625,
-0.03369140625,
0.02398681640625,
-0.030059814453125,
0.03350830078125,
0.03729248046875,
-0.018951416015625,
-0.007175445556640625,
-0.00905609130859375,
0.0165252685546875,
0.01580810546875,
-0.057159423828125,
-0.01959228515625,
0.0145721435546875,
0.0394287109375,
0.028411865234375,
0.054473876953125,
-0.003017425537109375,
0.0272216796875,
0.002407073974609375,
0.0312042236328125,
-0.00954437255859375,
0.00756072998046875,
-0.0306549072265625,
-0.006198883056640625,
-0.0031948089599609375,
-0.0160369873046875
]
] |
replit/replit-code-v1_5-3b | 2023-10-20T14:45:45.000Z | [
"transformers",
"pytorch",
"mpt",
"text-generation",
"code",
"Composer",
"MosaicML",
"llm-foundry",
"StreamingDatasets",
"custom_code",
"dataset:bigcode/the-stack-dedup",
"dataset:togethercomputer/RedPajama-Data-1T",
"license:apache-2.0",
"text-generation-inference",
"region:us"
] | text-generation | replit | null | null | replit/replit-code-v1_5-3b | 234 | 7,415 | transformers | 2023-10-09T18:59:50 | ---
license: apache-2.0
datasets:
- bigcode/the-stack-dedup
- togethercomputer/RedPajama-Data-1T
tags:
- code
- Composer
- MosaicML
- llm-foundry
- StreamingDatasets
language:
- code
---
# Replit Code V-1.5 3B
Developed by: Replit, Inc.
## Model Description
Replit Code v1.5 is a 3.3B parameter Causal Language Model focused on **Code Completion**.
The model is trained in `bfloat16` on 1T tokens of code (~200B tokens over 5 epochs, including linear cooldown) for 30 programming languages from a subset of permissively licensed code from Bigcode's [Stack Dedup dataset](https://huggingface.co/datasets/bigcode/the-stack-dedup), a filtered natural language sample from Markdown and reStructuredText subsets from the same Stack Dedup dataset, and a dev-oriented sample from [RedPajama's StackExchange dataset](https://github.com/togethercomputer/RedPajama-Data) sourced from the [Stack Exchange Data Dump by Stack Exchange Inc](https://archive.org/details/stackexchange).
The 30 programming languages are:
```
Java, JavaScript, C, PHP, Python, C++, C#, TypeScript, Go, CSS, HTML, Rust, Ruby, Swift, Scala, Shell, Lua, Perl, Haskell, JSX, Julia, Common Lisp, OCaml, Solidity, Scheme, R, Zig, SQL, Racket, D
```
The context size of the model is 4096 tokens. We use the GPTNeoX tokenizer with a custom trained and optimized vocabulary of 32768 tokens. This custom vocabulary led to single-digit % points on compression while maintaining or improving coverage on our training corpus.
The model has been trained on the [MosaicML](https://www.mosaicml.com/) platform on 128 H100-80GB GPUs using their [LLM Foundry](https://github.com/mosaicml/llm-foundry) and [Composer](https://github.com/mosaicml/composer) training library built on top of PyTorch.
## Dependencies
You will need to install the latest versions of the following dependencies:
```
einops
torch
transformers
```
## How to Use
### Generation
You can generate code using the `transformers` library as follows:
```python
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained('replit/replit-code-v1_5-3b', trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained('replit/replit-code-v1_5-3b', trust_remote_code=True)
x = tokenizer.encode('def fibonacci(n): ', return_tensors='pt')
y = model.generate(x, max_length=100, do_sample=True, top_p=0.95, top_k=4, temperature=0.2, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)
# decoding
generated_code = tokenizer.decode(y[0], skip_special_tokens=True, clean_up_tokenization_spaces=False)
print(generated_code)
```
Experiment with different decoding methods and parameters to get the best results for your use case.
### Using Triton Implementation of Flash Attention
```python
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, AutoConfig
config = AutoConfig.from_pretrained(
"replit/replit-code-v1_5-3b",
trust_remote_code=True
)
config.attn_config['attn_impl'] = 'triton'
# load model
tokenizer = AutoTokenizer.from_pretrained('replit/replit-code-v1_5-3b', trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained('replit/replit-code-v1_5-3b', config=config, trust_remote_code=True)
model.to(device='cuda:0', dtype=torch.bfloat16)
# forward pass
x = tokenizer.encode('def fibonacci(n): ', return_tensors='pt').to(device='cuda:0')
x = x.to(device='cuda:0')
y = model.generate(x, max_length=100, do_sample=True, top_p=0.95, top_k=4, temperature=0.2, num_return_sequences=1, eos_token_id=tokenizer.eos_token_id)
# decoding
generated_code = tokenizer.decode(y[0], skip_special_tokens=True, clean_up_tokenization_spaces=False)
print(generated_code)
```
Experiment with different decoding methods and parameters to get the best results for your use case. We recommend experimenting with `temperature` and `reptition_penalty`for optimal performance on your use case!
## Intended Use
Replit intends this model be used by anyone as a foundational model for application-specific fine-tuning without strict limitations on commercial use.
The model is trained specifically for code completion tasks.
## Limitations
The pre-training dataset may have contained offensive or inappropriate content even after applying data cleansing and toxicity and profanity filters, and such content may be reflected in model generated text. We recommend that users exercise reasonable caution when using in production systems. Do not use for any applications that may cause harm or distress to individuals or groups.
| 4,557 | [
[
-0.0226898193359375,
-0.0325927734375,
-0.002132415771484375,
0.028656005859375,
-0.0180206298828125,
0.00714111328125,
-0.01068115234375,
-0.0260009765625,
0.005359649658203125,
0.0291290283203125,
-0.033355712890625,
-0.05194091796875,
-0.0304107666015625,
0.0133209228515625,
-0.02825927734375,
0.06768798828125,
-0.0042266845703125,
-0.003185272216796875,
0.007793426513671875,
0.0031299591064453125,
-0.0073699951171875,
-0.040252685546875,
-0.043731689453125,
-0.010894775390625,
0.017242431640625,
0.0289459228515625,
0.0300445556640625,
0.031585693359375,
0.039276123046875,
0.027191162109375,
0.0035266876220703125,
0.00811004638671875,
-0.045806884765625,
-0.017669677734375,
0.0032520294189453125,
-0.043731689453125,
-0.031585693359375,
-0.004180908203125,
0.037994384765625,
0.0131988525390625,
0.00235748291015625,
0.044952392578125,
-0.025848388671875,
0.037109375,
-0.039794921875,
0.01323699951171875,
-0.04083251953125,
0.0018482208251953125,
0.006160736083984375,
-0.0253143310546875,
-0.042999267578125,
-0.043975830078125,
-0.0054473876953125,
-0.04156494140625,
0.0313720703125,
0.00009244680404663086,
0.07806396484375,
0.026580810546875,
-0.0155181884765625,
-0.01181793212890625,
-0.0372314453125,
0.0565185546875,
-0.07037353515625,
0.018463134765625,
0.0289459228515625,
0.0010137557983398438,
-0.0016727447509765625,
-0.086181640625,
-0.042266845703125,
-0.0189056396484375,
-0.0127410888671875,
0.006103515625,
-0.01678466796875,
-0.0017137527465820312,
0.037811279296875,
0.0263671875,
-0.048187255859375,
-0.01861572265625,
-0.04583740234375,
-0.019317626953125,
0.048065185546875,
0.01708984375,
0.0238037109375,
-0.0296783447265625,
-0.032989501953125,
-0.005382537841796875,
-0.042999267578125,
0.0131683349609375,
0.020782470703125,
-0.003597259521484375,
-0.0285186767578125,
0.031707763671875,
-0.01306915283203125,
0.053863525390625,
0.004924774169921875,
-0.0009613037109375,
0.0391845703125,
-0.036712646484375,
-0.03277587890625,
-0.0108184814453125,
0.06689453125,
0.01934814453125,
0.03033447265625,
-0.00850677490234375,
-0.0039520263671875,
-0.0113525390625,
0.01306915283203125,
-0.0609130859375,
-0.029205322265625,
0.03131103515625,
-0.0218963623046875,
-0.026641845703125,
0.01557159423828125,
-0.043670654296875,
0.0044708251953125,
-0.00390625,
0.038970947265625,
-0.04833984375,
-0.0234375,
0.029296875,
-0.017669677734375,
0.01776123046875,
-0.0008025169372558594,
-0.0732421875,
-0.00162506103515625,
0.033721923828125,
0.07025146484375,
0.0161285400390625,
-0.033111572265625,
-0.02508544921875,
0.002071380615234375,
-0.00409698486328125,
0.0215606689453125,
-0.012725830078125,
-0.01593017578125,
-0.00853729248046875,
0.00914764404296875,
-0.0140533447265625,
-0.014892578125,
0.00644683837890625,
-0.0513916015625,
0.01055145263671875,
0.002506256103515625,
-0.033447265625,
-0.0205535888671875,
0.0027942657470703125,
-0.050018310546875,
0.06884765625,
0.03961181640625,
-0.066650390625,
0.01293182373046875,
-0.041259765625,
-0.0272979736328125,
-0.004119873046875,
-0.004608154296875,
-0.046173095703125,
0.0008912086486816406,
0.01397705078125,
0.0192718505859375,
-0.0131988525390625,
0.00774383544921875,
-0.0131072998046875,
-0.042449951171875,
0.029205322265625,
-0.034027099609375,
0.0823974609375,
0.0300445556640625,
-0.044647216796875,
0.0151519775390625,
-0.06005859375,
0.01357269287109375,
0.00921630859375,
-0.0197601318359375,
0.01366424560546875,
-0.03387451171875,
0.0188446044921875,
0.0161285400390625,
0.036041259765625,
-0.038116455078125,
0.0235595703125,
-0.043609619140625,
0.055755615234375,
0.045379638671875,
0.0060272216796875,
0.0220947265625,
-0.017791748046875,
0.03887939453125,
0.005184173583984375,
0.024139404296875,
-0.0118255615234375,
-0.034332275390625,
-0.061431884765625,
-0.0216827392578125,
0.01898193359375,
0.03466796875,
-0.06390380859375,
0.03631591796875,
-0.01480865478515625,
-0.0460205078125,
-0.03997802734375,
-0.0011119842529296875,
0.032196044921875,
0.039093017578125,
0.034423828125,
0.00016689300537109375,
-0.0511474609375,
-0.055572509765625,
0.0045318603515625,
-0.0038280487060546875,
0.006389617919921875,
0.0094451904296875,
0.049468994140625,
-0.022430419921875,
0.07891845703125,
-0.03033447265625,
-0.0008044242858886719,
-0.0098419189453125,
0.0189056396484375,
0.038543701171875,
0.061248779296875,
0.046661376953125,
-0.056060791015625,
-0.033050537109375,
-0.0024261474609375,
-0.043792724609375,
0.00569915771484375,
-0.00974273681640625,
-0.0018405914306640625,
0.035125732421875,
0.02667236328125,
-0.0303497314453125,
0.0511474609375,
0.038238525390625,
-0.03582763671875,
0.038970947265625,
-0.02740478515625,
0.0222320556640625,
-0.092041015625,
0.0288848876953125,
-0.01396942138671875,
-0.017120361328125,
-0.043212890625,
0.0007123947143554688,
0.0205078125,
0.0029087066650390625,
-0.055419921875,
0.039886474609375,
-0.0275726318359375,
-0.00449371337890625,
-0.003631591796875,
-0.01134490966796875,
-0.00012201070785522461,
0.056854248046875,
0.0092620849609375,
0.06805419921875,
0.03900146484375,
-0.045562744140625,
0.039794921875,
0.013214111328125,
-0.024627685546875,
-0.0130157470703125,
-0.055145263671875,
0.00922393798828125,
0.0195770263671875,
0.00957489013671875,
-0.090576171875,
-0.0142669677734375,
0.0206451416015625,
-0.0552978515625,
0.0143890380859375,
-0.0156402587890625,
-0.041778564453125,
-0.05322265625,
-0.01348876953125,
0.049102783203125,
0.04620361328125,
-0.040863037109375,
0.026336669921875,
0.00809478759765625,
0.0233001708984375,
-0.055572509765625,
-0.0732421875,
-0.01371002197265625,
-0.00640869140625,
-0.05255126953125,
0.0246429443359375,
-0.00519561767578125,
0.01171875,
-0.01241302490234375,
-0.0094757080078125,
-0.0017023086547851562,
0.0088348388671875,
0.0158843994140625,
0.027099609375,
-0.00951385498046875,
0.0003094673156738281,
-0.01812744140625,
-0.02691650390625,
0.0012903213500976562,
-0.027984619140625,
0.07720947265625,
-0.03826904296875,
-0.01922607421875,
-0.042877197265625,
0.0012664794921875,
0.050537109375,
-0.02490234375,
0.05474853515625,
0.0731201171875,
-0.028564453125,
0.0003826618194580078,
-0.03814697265625,
-0.00443267822265625,
-0.040924072265625,
0.025634765625,
-0.024078369140625,
-0.06500244140625,
0.061614990234375,
0.025665283203125,
0.005298614501953125,
0.035125732421875,
0.05108642578125,
0.0177001953125,
0.0611572265625,
0.03778076171875,
-0.005889892578125,
0.04656982421875,
-0.06951904296875,
0.01165008544921875,
-0.0562744140625,
-0.01113128662109375,
-0.025482177734375,
-0.01406097412109375,
-0.03851318359375,
-0.0240020751953125,
0.0129241943359375,
0.01557159423828125,
-0.04583740234375,
0.05902099609375,
-0.06256103515625,
0.0233001708984375,
0.05633544921875,
0.0165252685546875,
0.0081787109375,
0.004726409912109375,
0.00037479400634765625,
0.0143585205078125,
-0.058807373046875,
-0.0246734619140625,
0.08929443359375,
0.0347900390625,
0.0572509765625,
-0.0035552978515625,
0.06494140625,
-0.0001283884048461914,
0.0278778076171875,
-0.0303497314453125,
0.0244140625,
0.0011739730834960938,
-0.05133056640625,
-0.0197601318359375,
-0.04248046875,
-0.05230712890625,
-0.00345611572265625,
0.01120758056640625,
-0.06805419921875,
0.02801513671875,
0.006969451904296875,
-0.047607421875,
0.0270233154296875,
-0.069580078125,
0.08184814453125,
-0.0154571533203125,
-0.0308685302734375,
-0.0083770751953125,
-0.04266357421875,
0.0276031494140625,
0.00860595703125,
-0.0031986236572265625,
0.006866455078125,
0.003276824951171875,
0.0595703125,
-0.0291290283203125,
0.05743408203125,
-0.0023746490478515625,
-0.0035190582275390625,
0.02288818359375,
-0.0008916854858398438,
0.0280609130859375,
0.03643798828125,
-0.013885498046875,
0.025970458984375,
0.0231170654296875,
-0.034515380859375,
-0.0190277099609375,
0.0501708984375,
-0.07135009765625,
-0.03424072265625,
-0.032928466796875,
-0.04376220703125,
0.0067138671875,
0.022705078125,
0.059356689453125,
0.045074462890625,
0.0161285400390625,
0.00702667236328125,
0.037750244140625,
-0.0231170654296875,
0.0634765625,
0.03961181640625,
-0.027374267578125,
-0.055145263671875,
0.07598876953125,
-0.005390167236328125,
0.0304107666015625,
0.015045166015625,
-0.0095672607421875,
-0.0158233642578125,
-0.033660888671875,
-0.02191162109375,
0.0212554931640625,
-0.047607421875,
-0.041046142578125,
-0.038665771484375,
-0.033966064453125,
-0.054840087890625,
-0.01806640625,
-0.03338623046875,
-0.0151214599609375,
-0.045806884765625,
-0.0029468536376953125,
0.04937744140625,
0.0310821533203125,
-0.01238250732421875,
0.02972412109375,
-0.06182861328125,
0.0293426513671875,
0.00970458984375,
0.036529541015625,
-0.0064849853515625,
-0.05035400390625,
-0.03900146484375,
0.0088348388671875,
-0.00540924072265625,
-0.062469482421875,
0.048187255859375,
-0.00807952880859375,
0.028717041015625,
0.022247314453125,
0.006603240966796875,
0.04754638671875,
-0.01517486572265625,
0.0623779296875,
0.0205841064453125,
-0.0831298828125,
0.03814697265625,
-0.01277923583984375,
0.031402587890625,
0.037109375,
0.0186614990234375,
-0.038238525390625,
-0.0321044921875,
-0.06353759765625,
-0.06195068359375,
0.08294677734375,
0.026641845703125,
0.001232147216796875,
-0.0191192626953125,
0.0279693603515625,
-0.00736236572265625,
0.0159912109375,
-0.07061767578125,
-0.036865234375,
-0.032623291015625,
-0.0243988037109375,
0.003154754638671875,
-0.0063018798828125,
-0.006557464599609375,
-0.03887939453125,
0.0438232421875,
0.004787445068359375,
0.045684814453125,
0.03460693359375,
-0.02777099609375,
-0.0169219970703125,
-0.00968170166015625,
0.06640625,
0.0601806640625,
-0.0300750732421875,
-0.006534576416015625,
0.01422119140625,
-0.05352783203125,
0.004016876220703125,
0.01334381103515625,
-0.00774383544921875,
0.0032215118408203125,
0.0350341796875,
0.056304931640625,
0.018463134765625,
-0.026275634765625,
0.029541015625,
-0.0100860595703125,
-0.0130462646484375,
-0.03460693359375,
0.0172119140625,
0.0026836395263671875,
0.018768310546875,
0.03887939453125,
0.0073089599609375,
-0.0128326416015625,
-0.024169921875,
0.037994384765625,
0.01277923583984375,
-0.0131988525390625,
-0.0149078369140625,
0.057159423828125,
0.01129913330078125,
-0.02899169921875,
0.046905517578125,
-0.0151214599609375,
-0.05047607421875,
0.0806884765625,
0.052734375,
0.07061767578125,
-0.0031948089599609375,
0.0171661376953125,
0.045928955078125,
0.03155517578125,
0.006443023681640625,
0.0301361083984375,
-0.0011835098266601562,
-0.042572021484375,
-0.018218994140625,
-0.05010986328125,
-0.004222869873046875,
-0.005428314208984375,
-0.038360595703125,
0.025909423828125,
-0.044830322265625,
-0.014404296875,
-0.0233612060546875,
0.01480865478515625,
-0.061431884765625,
0.00923919677734375,
0.0190277099609375,
0.0836181640625,
-0.046661376953125,
0.0777587890625,
0.032745361328125,
-0.06573486328125,
-0.0726318359375,
-0.0164337158203125,
-0.030609130859375,
-0.072265625,
0.058807373046875,
0.01434326171875,
0.00013172626495361328,
0.038726806640625,
-0.0472412109375,
-0.08251953125,
0.10272216796875,
0.0266876220703125,
-0.0357666015625,
-0.0028629302978515625,
0.01177978515625,
0.042510986328125,
-0.034515380859375,
0.041473388671875,
0.04522705078125,
0.0275726318359375,
0.006763458251953125,
-0.065185546875,
0.005268096923828125,
-0.01428985595703125,
-0.0035381317138671875,
0.0024662017822265625,
-0.061737060546875,
0.0831298828125,
-0.0217437744140625,
-0.0218963623046875,
0.0102081298828125,
0.054962158203125,
0.01849365234375,
0.0150909423828125,
0.00917816162109375,
0.044189453125,
0.042388916015625,
-0.01157379150390625,
0.07537841796875,
-0.0572509765625,
0.072998046875,
0.06573486328125,
0.014312744140625,
0.049560546875,
0.031829833984375,
-0.0146484375,
0.03533935546875,
0.0517578125,
-0.01763916015625,
0.038787841796875,
0.0127410888671875,
-0.005603790283203125,
-0.004657745361328125,
0.0291290283203125,
-0.05047607421875,
0.0080108642578125,
0.026336669921875,
-0.034027099609375,
-0.01108551025390625,
0.0078887939453125,
-0.00936126708984375,
-0.0269317626953125,
-0.0213775634765625,
0.030242919921875,
0.0016765594482421875,
-0.03717041015625,
0.07305908203125,
-0.0012273788452148438,
0.05755615234375,
-0.041412353515625,
0.006992340087890625,
-0.025054931640625,
0.0157928466796875,
-0.033111572265625,
-0.050567626953125,
0.0181884765625,
-0.00994873046875,
-0.00579071044921875,
-0.002552032470703125,
0.030181884765625,
-0.037994384765625,
-0.03546142578125,
0.0098114013671875,
-0.00568389892578125,
0.0023212432861328125,
0.00505828857421875,
-0.054840087890625,
0.00878143310546875,
0.0186614990234375,
-0.0325927734375,
-0.00835418701171875,
0.020355224609375,
0.028839111328125,
0.055938720703125,
0.046844482421875,
-0.00656890869140625,
0.027008056640625,
0.00370025634765625,
0.06610107421875,
-0.0634765625,
-0.033172607421875,
-0.0611572265625,
0.04864501953125,
-0.001499176025390625,
-0.049896240234375,
0.058624267578125,
0.051361083984375,
0.06988525390625,
-0.025238037109375,
0.03887939453125,
-0.0208892822265625,
0.0068206787109375,
-0.036712646484375,
0.06451416015625,
-0.0285186767578125,
0.0218658447265625,
-0.0164642333984375,
-0.05682373046875,
-0.0224609375,
0.052734375,
-0.036651611328125,
0.007053375244140625,
0.056427001953125,
0.0697021484375,
0.0015974044799804688,
-0.0027751922607421875,
0.0165863037109375,
0.0121307373046875,
0.02655029296875,
0.071044921875,
0.037384033203125,
-0.0667724609375,
0.046722412109375,
-0.037017822265625,
-0.0114898681640625,
-0.0272064208984375,
-0.044921875,
-0.0726318359375,
-0.041717529296875,
-0.0226287841796875,
-0.041168212890625,
-0.0211944580078125,
0.0765380859375,
0.0482177734375,
-0.06744384765625,
-0.00821685791015625,
-0.0296630859375,
0.018890380859375,
-0.008758544921875,
-0.02191162109375,
0.035247802734375,
-0.045196533203125,
-0.055023193359375,
-0.00531005859375,
0.01751708984375,
0.00174713134765625,
-0.025238037109375,
-0.011688232421875,
-0.029296875,
-0.00850677490234375,
0.0201873779296875,
0.03607177734375,
-0.051177978515625,
-0.0222625732421875,
0.00794219970703125,
-0.03533935546875,
0.01029205322265625,
0.05029296875,
-0.05157470703125,
0.0277099609375,
0.04913330078125,
0.0297088623046875,
0.04766845703125,
-0.0058441162109375,
0.0357666015625,
-0.04638671875,
0.0215301513671875,
0.014617919921875,
0.0308837890625,
0.00432586669921875,
-0.041290283203125,
0.042510986328125,
0.0267791748046875,
-0.043975830078125,
-0.0584716796875,
0.00693511962890625,
-0.07904052734375,
-0.0181427001953125,
0.09234619140625,
-0.01557159423828125,
-0.04180908203125,
-0.0012760162353515625,
-0.01435089111328125,
0.03973388671875,
-0.018035888671875,
0.050750732421875,
0.037445068359375,
-0.005126953125,
-0.0194854736328125,
-0.0215301513671875,
0.0482177734375,
0.0219573974609375,
-0.03826904296875,
0.005695343017578125,
0.01849365234375,
0.035308837890625,
0.02789306640625,
0.044830322265625,
-0.00852203369140625,
0.05255126953125,
0.0036678314208984375,
0.02886962890625,
-0.033660888671875,
-0.010498046875,
-0.032470703125,
0.00745391845703125,
-0.024871826171875,
-0.025634765625
]
] |
zarakiquemparte/zararp-1.1-l2-7b | 2023-09-14T12:37:12.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama2",
"license:other",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | zarakiquemparte | null | null | zarakiquemparte/zararp-1.1-l2-7b | 2 | 7,413 | transformers | 2023-09-14T11:05:31 | ---
license: other
tags:
- llama2
---
# Model Card: ZaraRP 1.1 L2 7b
This model uses [Nous Hermes Llama2 7b](https://huggingface.co/NousResearch/Nous-Hermes-llama-2-7b) (53%) as a base with [Stable Beluga 7b](https://huggingface.co/stabilityai/StableBeluga-7B) (47%) and the result of this merge was merged with [LimaRP Llama2 v2 Lora 7b](https://huggingface.co/lemonilia/limarp-llama2-v2) and [PIPPA ShareGPT Subset Variation Two Lora 7b](https://huggingface.co/zarakiquemparte/PIPPA-ShareGPT-Subset-Lora-VT-7b).
This merge of models(hermes and stable beluga) was done with this [script](https://github.com/zarakiquemparte/zaraki-tools/blob/main/merge-cli.py)
This merge of Lora with Model was done with this [script](https://github.com/zarakiquemparte/zaraki-tools/blob/main/apply-lora.py)
Merge illustration:
## Usage:
Since this is a merge between Nous Hermes, Stable Beluga, LimaRP, and PIPPA ShareGPT, the following instruction formats should work:
Alpaca 2:
```
### Instruction:
<prompt>
### Response:
<leave a newline blank for model to respond>
```
Custom:
```
SYSTEM: Do thing
USER: {prompt}
CHARACTER:
```
Alpaca LimaRP instruction format:
```
### 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}
```
## Bias, Risks, and Limitations
This model is not intended for supplying factual information or advice in any form
## Training Details
This model is merged and can be reproduced using the tools mentioned above. Please refer to all provided links for extra model-specific details. | 1,919 | [
[
-0.03900146484375,
-0.03997802734375,
0.0175628662109375,
0.03302001953125,
-0.0301513671875,
-0.0248565673828125,
0.0248565673828125,
-0.04962158203125,
0.0341796875,
0.06683349609375,
-0.06414794921875,
-0.0283050537109375,
-0.037750244140625,
-0.0104827880859375,
-0.025543212890625,
0.09375,
0.0155792236328125,
-0.0204010009765625,
0.0155181884765625,
-0.0191497802734375,
-0.04400634765625,
-0.03802490234375,
-0.05609130859375,
-0.04296875,
0.052490234375,
0.0290679931640625,
0.054412841796875,
0.028106689453125,
0.033203125,
0.0281982421875,
-0.030670166015625,
0.01302337646484375,
-0.025238037109375,
0.0014047622680664062,
-0.007457733154296875,
-0.035736083984375,
-0.0748291015625,
0.0008454322814941406,
0.0308380126953125,
0.032073974609375,
-0.0207977294921875,
0.01702880859375,
-0.005084991455078125,
0.02777099609375,
-0.0333251953125,
-0.0027294158935546875,
-0.01422882080078125,
0.01059722900390625,
0.0012569427490234375,
-0.0017404556274414062,
-0.01384735107421875,
-0.0300445556640625,
0.01085662841796875,
-0.04620361328125,
0.0009646415710449219,
-0.0166168212890625,
0.07391357421875,
0.0275421142578125,
-0.03155517578125,
-0.01403045654296875,
-0.044647216796875,
0.03997802734375,
-0.07061767578125,
-0.00612640380859375,
0.025665283203125,
0.0301513671875,
-0.025360107421875,
-0.037750244140625,
-0.05426025390625,
0.00018680095672607422,
0.0019683837890625,
0.00502777099609375,
-0.035675048828125,
-0.01264190673828125,
0.031890869140625,
0.0248870849609375,
-0.0241546630859375,
0.0193328857421875,
-0.06585693359375,
0.0058746337890625,
0.045806884765625,
0.01544189453125,
0.0157623291015625,
-0.022308349609375,
-0.045684814453125,
-0.032928466796875,
-0.035247802734375,
0.003726959228515625,
0.04669189453125,
0.0238189697265625,
-0.059844970703125,
0.07720947265625,
-0.0101165771484375,
0.050811767578125,
0.023773193359375,
-0.017120361328125,
0.039764404296875,
0.0020275115966796875,
-0.03790283203125,
0.0002830028533935547,
0.06683349609375,
0.0382080078125,
-0.0171661376953125,
0.0257110595703125,
0.00045037269592285156,
0.0026702880859375,
0.00005936622619628906,
-0.055694580078125,
0.01267242431640625,
0.037109375,
-0.031829833984375,
-0.0328369140625,
-0.01061248779296875,
-0.053070068359375,
-0.034393310546875,
0.00421142578125,
0.0244140625,
-0.029510498046875,
-0.0183258056640625,
0.0211181640625,
-0.010467529296875,
0.053436279296875,
0.0303497314453125,
-0.060821533203125,
0.0213775634765625,
0.03692626953125,
0.047027587890625,
0.0272369384765625,
-0.01995849609375,
-0.035797119140625,
0.00873565673828125,
-0.0203399658203125,
0.05914306640625,
-0.00439453125,
-0.04302978515625,
-0.0099029541015625,
0.005924224853515625,
-0.0018110275268554688,
-0.031982421875,
0.05853271484375,
-0.0260467529296875,
0.03594970703125,
-0.0254364013671875,
-0.019195556640625,
-0.03912353515625,
0.021514892578125,
-0.058837890625,
0.07427978515625,
0.0206756591796875,
-0.06756591796875,
-0.006702423095703125,
-0.07159423828125,
-0.0164031982421875,
-0.00760650634765625,
0.01094818115234375,
-0.035064697265625,
-0.00498199462890625,
-0.0012331008911132812,
0.033905029296875,
-0.022613525390625,
-0.01206207275390625,
-0.031585693359375,
-0.0251312255859375,
0.0177764892578125,
0.002582550048828125,
0.06964111328125,
0.01885986328125,
-0.009979248046875,
0.0028476715087890625,
-0.06304931640625,
0.00539398193359375,
0.0213165283203125,
-0.02703857421875,
-0.0004763603210449219,
-0.016082763671875,
0.0111236572265625,
0.01178741455078125,
0.035430908203125,
-0.0208282470703125,
0.050384521484375,
-0.0252838134765625,
0.01336669921875,
0.0307769775390625,
-0.0014486312866210938,
0.01702880859375,
-0.050445556640625,
0.030059814453125,
-0.0008082389831542969,
0.019561767578125,
-0.002353668212890625,
-0.0615234375,
-0.0838623046875,
-0.0225830078125,
0.012115478515625,
0.04638671875,
-0.0272216796875,
0.0633544921875,
0.0005159378051757812,
-0.056610107421875,
-0.03271484375,
-0.0126800537109375,
0.02178955078125,
0.03424072265625,
0.0211181640625,
-0.034942626953125,
-0.05780029296875,
-0.07501220703125,
0.02447509765625,
-0.0257110595703125,
-0.00719451904296875,
0.006046295166015625,
0.0411376953125,
-0.04498291015625,
0.0540771484375,
-0.030364990234375,
-0.0227813720703125,
-0.047393798828125,
0.0258636474609375,
0.044036865234375,
0.0504150390625,
0.05908203125,
-0.037567138671875,
-0.0226898193359375,
0.0036029815673828125,
-0.051025390625,
-0.00815582275390625,
0.0122528076171875,
-0.0062713623046875,
0.02178955078125,
-0.00870513916015625,
-0.06524658203125,
0.052734375,
0.054595947265625,
-0.038787841796875,
0.01381683349609375,
-0.031768798828125,
0.0196075439453125,
-0.10400390625,
0.02886962890625,
-0.01140594482421875,
-0.0015964508056640625,
-0.054931640625,
0.0230255126953125,
-0.0117645263671875,
-0.00569915771484375,
-0.04156494140625,
0.05810546875,
-0.0219268798828125,
-0.020294189453125,
-0.033538818359375,
-0.00655364990234375,
0.004703521728515625,
0.0391845703125,
0.0087738037109375,
0.0221710205078125,
0.0269012451171875,
-0.042816162109375,
0.04156494140625,
0.049072265625,
-0.0185546875,
0.035736083984375,
-0.06561279296875,
0.0204925537109375,
0.003978729248046875,
0.0302734375,
-0.06414794921875,
-0.0297088623046875,
0.060821533203125,
-0.0266876220703125,
0.0042572021484375,
-0.0273895263671875,
-0.042083740234375,
-0.023223876953125,
-0.0203704833984375,
0.01366424560546875,
0.060882568359375,
-0.040069580078125,
0.066162109375,
-0.003383636474609375,
-0.0012454986572265625,
-0.052490234375,
-0.0584716796875,
-0.0167694091796875,
-0.0306549072265625,
-0.0615234375,
0.016632080078125,
-0.0249481201171875,
-0.006130218505859375,
-0.01139068603515625,
0.01314544677734375,
-0.0206298828125,
-0.006641387939453125,
0.034759521484375,
0.04522705078125,
-0.01128387451171875,
-0.0303497314453125,
0.0096893310546875,
-0.0024662017822265625,
-0.0253448486328125,
0.0352783203125,
0.070068359375,
-0.002338409423828125,
-0.0288543701171875,
-0.047821044921875,
0.0265655517578125,
0.057830810546875,
-0.0171966552734375,
0.07489013671875,
0.041717529296875,
-0.0171356201171875,
0.01306915283203125,
-0.053985595703125,
0.0107574462890625,
-0.0302734375,
0.0126800537109375,
-0.026458740234375,
-0.0253143310546875,
0.072265625,
0.02264404296875,
0.01059722900390625,
0.042724609375,
0.038726806640625,
0.0019931793212890625,
0.06036376953125,
0.06719970703125,
-0.0128326416015625,
0.0286407470703125,
-0.0467529296875,
0.01355743408203125,
-0.07342529296875,
-0.0391845703125,
-0.039306640625,
-0.024200439453125,
-0.037750244140625,
-0.037506103515625,
0.00738525390625,
0.0238189697265625,
-0.0264129638671875,
0.05072021484375,
-0.023895263671875,
0.0218353271484375,
0.033477783203125,
0.0112152099609375,
0.022674560546875,
0.00730133056640625,
0.019012451171875,
0.01556396484375,
-0.04876708984375,
-0.0343017578125,
0.0706787109375,
0.037261962890625,
0.057952880859375,
0.0255126953125,
0.06640625,
0.00811004638671875,
0.009063720703125,
-0.041259765625,
0.048614501953125,
0.0046539306640625,
-0.038299560546875,
-0.005947113037109375,
-0.01276397705078125,
-0.06756591796875,
0.02337646484375,
-0.021331787109375,
-0.0645751953125,
0.02239990234375,
0.001125335693359375,
-0.057342529296875,
0.0152587890625,
-0.04937744140625,
0.056915283203125,
-0.00687408447265625,
-0.01360321044921875,
-0.0108489990234375,
-0.051727294921875,
0.04498291015625,
0.00817108154296875,
-0.006992340087890625,
-0.01108551025390625,
-0.004131317138671875,
0.056121826171875,
-0.04022216796875,
0.074462890625,
0.00943756103515625,
-0.0280609130859375,
0.0263519287109375,
-0.004291534423828125,
0.0411376953125,
-0.0214080810546875,
-0.002307891845703125,
0.01180267333984375,
0.0060577392578125,
-0.0178680419921875,
-0.036590576171875,
0.048675537109375,
-0.078857421875,
-0.0263519287109375,
-0.045867919921875,
-0.038970947265625,
0.00737762451171875,
-0.004756927490234375,
0.030731201171875,
0.0160980224609375,
-0.0128021240234375,
-0.00627899169921875,
0.0469970703125,
-0.0200042724609375,
0.0237884521484375,
0.055877685546875,
-0.0244293212890625,
-0.0278167724609375,
0.032012939453125,
-0.019683837890625,
0.01227569580078125,
0.006378173828125,
0.004947662353515625,
-0.0206756591796875,
-0.006641387939453125,
-0.027740478515625,
0.045166015625,
-0.054779052734375,
-0.0141754150390625,
-0.037628173828125,
-0.037353515625,
-0.030670166015625,
-0.00209808349609375,
-0.0273284912109375,
-0.03363037109375,
-0.025390625,
0.006229400634765625,
0.039947509765625,
0.05523681640625,
-0.0226287841796875,
0.05450439453125,
-0.06182861328125,
0.02777099609375,
0.022674560546875,
-0.0020465850830078125,
-0.00971221923828125,
-0.0650634765625,
0.00983428955078125,
0.01311492919921875,
-0.02752685546875,
-0.080810546875,
0.051513671875,
-0.009552001953125,
0.036834716796875,
0.033599853515625,
-0.0014810562133789062,
0.055694580078125,
-0.00829315185546875,
0.049774169921875,
0.031524658203125,
-0.05682373046875,
0.052978515625,
-0.039459228515625,
0.01099395751953125,
0.01227569580078125,
0.0303802490234375,
-0.048553466796875,
-0.001873016357421875,
-0.061126708984375,
-0.04302978515625,
0.06756591796875,
0.0224151611328125,
0.007228851318359375,
0.0189666748046875,
0.0178070068359375,
-0.0138092041015625,
0.014739990234375,
-0.0726318359375,
-0.028106689453125,
-0.0037326812744140625,
-0.0016164779663085938,
-0.00015282630920410156,
-0.027008056640625,
-0.034942626953125,
-0.008331298828125,
0.051727294921875,
0.0037384033203125,
0.016998291015625,
0.00299072265625,
0.0271148681640625,
-0.020477294921875,
0.016448974609375,
0.052093505859375,
0.018829345703125,
-0.03936767578125,
-0.03106689453125,
0.0126190185546875,
-0.00945281982421875,
-0.0038089752197265625,
0.0232696533203125,
-0.00246429443359375,
-0.011260986328125,
0.0330810546875,
0.061737060546875,
0.01410675048828125,
-0.041595458984375,
0.029815673828125,
-0.0102386474609375,
-0.021514892578125,
-0.01445770263671875,
0.0172576904296875,
0.0100555419921875,
0.04644775390625,
0.0032444000244140625,
0.019622802734375,
-0.0033397674560546875,
-0.06561279296875,
-0.0298309326171875,
0.0235443115234375,
0.0019741058349609375,
-0.0179595947265625,
0.037353515625,
0.0168304443359375,
-0.025634765625,
0.045135498046875,
0.0008511543273925781,
-0.0303192138671875,
0.0516357421875,
0.040130615234375,
0.049713134765625,
-0.0325927734375,
0.01239013671875,
0.03912353515625,
0.006591796875,
-0.021392822265625,
0.031829833984375,
0.001766204833984375,
-0.04681396484375,
-0.017669677734375,
-0.03167724609375,
-0.0283660888671875,
0.039764404296875,
-0.049560546875,
0.040985107421875,
-0.03857421875,
-0.0140380859375,
-0.0157928466796875,
0.0159149169921875,
-0.03399658203125,
0.005828857421875,
0.01335906982421875,
0.064208984375,
-0.07861328125,
0.0655517578125,
0.04522705078125,
-0.051605224609375,
-0.07183837890625,
-0.0214385986328125,
-0.01114654541015625,
-0.064453125,
0.05267333984375,
0.0061798095703125,
0.0031986236572265625,
-0.038360595703125,
-0.045654296875,
-0.06768798828125,
0.087646484375,
0.0330810546875,
-0.0303802490234375,
-0.0033588409423828125,
-0.003170013427734375,
0.034332275390625,
-0.0330810546875,
0.030059814453125,
0.02996826171875,
0.02752685546875,
0.054107666015625,
-0.089111328125,
-0.00772857666015625,
-0.024627685546875,
-0.00933074951171875,
-0.0109100341796875,
-0.0650634765625,
0.08807373046875,
-0.0234527587890625,
0.0013761520385742188,
0.070556640625,
0.04840087890625,
0.04827880859375,
0.0015878677368164062,
0.0282135009765625,
0.06036376953125,
0.043731689453125,
0.003879547119140625,
0.06695556640625,
-0.0129241943359375,
0.029754638671875,
0.08642578125,
-0.034393310546875,
0.08062744140625,
0.030731201171875,
0.005794525146484375,
0.05389404296875,
0.050018310546875,
-0.0104827880859375,
0.03607177734375,
0.00133514404296875,
-0.0164947509765625,
-0.00916290283203125,
-0.0013151168823242188,
-0.06304931640625,
0.040618896484375,
0.00983428955078125,
-0.024749755859375,
-0.012054443359375,
-0.031890869140625,
-0.0019369125366210938,
-0.01983642578125,
-0.0160369873046875,
0.033538818359375,
-0.01372528076171875,
-0.056304931640625,
0.046905517578125,
0.0161590576171875,
0.05499267578125,
-0.0782470703125,
-0.02374267578125,
-0.045867919921875,
0.023406982421875,
-0.02191162109375,
-0.04052734375,
0.00988006591796875,
0.003490447998046875,
-0.0169525146484375,
0.0150299072265625,
0.041259765625,
-0.0227508544921875,
-0.045562744140625,
0.0265960693359375,
0.0294189453125,
0.027862548828125,
0.0194549560546875,
-0.05523681640625,
0.036712646484375,
0.0018987655639648438,
-0.01042938232421875,
0.027435302734375,
0.0106201171875,
0.0122222900390625,
0.07220458984375,
0.036651611328125,
-0.00479888916015625,
-0.00481414794921875,
-0.0017910003662109375,
0.07257080078125,
-0.031707763671875,
-0.03692626953125,
-0.036529541015625,
0.047637939453125,
-0.007343292236328125,
-0.02374267578125,
0.048980712890625,
0.044830322265625,
0.031829833984375,
-0.0115966796875,
0.04931640625,
-0.022705078125,
0.040771484375,
-0.040313720703125,
0.048980712890625,
-0.044342041015625,
0.0249176025390625,
-0.0192108154296875,
-0.07281494140625,
0.0080108642578125,
0.07012939453125,
-0.0002307891845703125,
0.004856109619140625,
0.0254058837890625,
0.057708740234375,
-0.017333984375,
-0.0208587646484375,
0.0211944580078125,
0.0116119384765625,
0.005970001220703125,
0.0557861328125,
0.0726318359375,
-0.066162109375,
0.02764892578125,
-0.011566162109375,
-0.013946533203125,
-0.0191497802734375,
-0.06829833984375,
-0.0849609375,
-0.0241241455078125,
-0.02337646484375,
-0.032379150390625,
-0.0050506591796875,
0.0723876953125,
0.059326171875,
-0.033660888671875,
-0.04193115234375,
0.02117919921875,
-0.003963470458984375,
-0.0010662078857421875,
-0.00971221923828125,
0.0137786865234375,
0.0171356201171875,
-0.06427001953125,
0.02691650390625,
-0.004425048828125,
0.0487060546875,
-0.0145721435546875,
-0.0216217041015625,
-0.01346588134765625,
0.01026153564453125,
0.0251922607421875,
0.052825927734375,
-0.06103515625,
-0.00258636474609375,
-0.017578125,
-0.0018787384033203125,
0.0008187294006347656,
0.0300445556640625,
-0.050628662109375,
-0.00843048095703125,
0.0271453857421875,
0.00304412841796875,
0.043060302734375,
-0.004825592041015625,
0.036529541015625,
-0.042724609375,
0.035003662109375,
-0.01348114013671875,
0.04241943359375,
0.0251922607421875,
-0.015899658203125,
0.048065185546875,
0.0005130767822265625,
-0.0250244140625,
-0.05523681640625,
0.00803375244140625,
-0.1143798828125,
-0.00421905517578125,
0.07391357421875,
-0.0000035762786865234375,
-0.0258026123046875,
0.0277252197265625,
-0.04949951171875,
0.01189422607421875,
-0.0305328369140625,
0.0445556640625,
0.026763916015625,
-0.0197601318359375,
-0.01471710205078125,
-0.01320648193359375,
0.01299285888671875,
0.005672454833984375,
-0.05950927734375,
-0.012359619140625,
0.0260009765625,
0.03558349609375,
0.034393310546875,
0.0504150390625,
0.00970458984375,
0.0253143310546875,
-0.00750732421875,
0.032440185546875,
-0.00687408447265625,
-0.00420379638671875,
-0.019317626953125,
-0.0095977783203125,
-0.00809478759765625,
-0.0169525146484375
]
] |
timm/efficientnet_b5.sw_in12k_ft_in1k | 2023-04-27T21:11:43.000Z | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"dataset:imagenet-1k",
"dataset:imagenet-12k",
"arxiv:1905.11946",
"license:apache-2.0",
"region:us"
] | image-classification | timm | null | null | timm/efficientnet_b5.sw_in12k_ft_in1k | 2 | 7,412 | timm | 2022-12-12T23:57:29 | ---
tags:
- image-classification
- timm
library_name: timm
license: apache-2.0
datasets:
- imagenet-1k
- imagenet-12k
---
# Model card for efficientnet_b5.sw_in12k_ft_in1k
A EfficientNet image classification model. Pretrained on ImageNet-12k and fine-tuned on ImageNet-1k by Ross Wightman in `timm` using recipe template described below.
Recipe details:
* Based on Swin Transformer train / pretrain recipe with modifications (related to both DeiT and ConvNeXt recipes)
* AdamW optimizer, gradient clipping, EMA weight averaging
* Cosine LR schedule with warmup
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 30.4
- GMACs: 9.6
- Activations (M): 93.6
- Image size: 448 x 448
- **Papers:**
- EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks: https://arxiv.org/abs/1905.11946
- **Dataset:** ImageNet-1k
- **Pretrain Dataset:** ImageNet-12k
- **Original:** https://github.com/huggingface/pytorch-image-models
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('efficientnet_b5.sw_in12k_ft_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Feature Map Extraction
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'efficientnet_b5.sw_in12k_ft_in1k',
pretrained=True,
features_only=True,
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
for o in output:
# print shape of each feature map in output
# e.g.:
# torch.Size([1, 24, 224, 224])
# torch.Size([1, 40, 112, 112])
# torch.Size([1, 64, 56, 56])
# torch.Size([1, 176, 28, 28])
# torch.Size([1, 512, 14, 14])
print(o.shape)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'efficientnet_b5.sw_in12k_ft_in1k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 2048, 14, 14) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
## Citation
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}
```
```bibtex
@inproceedings{tan2019efficientnet,
title={Efficientnet: Rethinking model scaling for convolutional neural networks},
author={Tan, Mingxing and Le, Quoc},
booktitle={International conference on machine learning},
pages={6105--6114},
year={2019},
organization={PMLR}
}
```
| 4,402 | [
[
-0.032958984375,
-0.03680419921875,
-0.007274627685546875,
0.0114288330078125,
-0.0196380615234375,
-0.033538818359375,
-0.0213775634765625,
-0.031951904296875,
0.019012451171875,
0.0267333984375,
-0.036590576171875,
-0.0450439453125,
-0.055633544921875,
-0.009613037109375,
-0.01451873779296875,
0.06744384765625,
-0.004619598388671875,
0.00516510009765625,
-0.00917816162109375,
-0.04498291015625,
-0.00867462158203125,
-0.017181396484375,
-0.0679931640625,
-0.0386962890625,
0.029632568359375,
0.0211181640625,
0.041473388671875,
0.05108642578125,
0.055145263671875,
0.03900146484375,
-0.00782012939453125,
0.004726409912109375,
-0.022918701171875,
-0.01354217529296875,
0.033355712890625,
-0.043212890625,
-0.034881591796875,
0.01898193359375,
0.051422119140625,
0.034881591796875,
-0.0025806427001953125,
0.034942626953125,
0.00901031494140625,
0.044403076171875,
-0.0221405029296875,
0.00806427001953125,
-0.03668212890625,
0.01023101806640625,
-0.00778961181640625,
0.0118560791015625,
-0.0208282470703125,
-0.0283660888671875,
0.0165557861328125,
-0.04315185546875,
0.034088134765625,
0.0006546974182128906,
0.10015869140625,
0.019683837890625,
-0.0012493133544921875,
-0.006725311279296875,
-0.01274871826171875,
0.05584716796875,
-0.0635986328125,
0.0176239013671875,
0.01419830322265625,
0.016937255859375,
-0.002689361572265625,
-0.0838623046875,
-0.03985595703125,
-0.0170135498046875,
-0.0174102783203125,
-0.004669189453125,
-0.023468017578125,
0.006519317626953125,
0.0301971435546875,
0.024444580078125,
-0.033050537109375,
0.0115814208984375,
-0.035064697265625,
-0.0156402587890625,
0.0418701171875,
0.0020465850830078125,
0.0266571044921875,
-0.0110626220703125,
-0.037139892578125,
-0.034393310546875,
-0.0297088623046875,
0.025115966796875,
0.0233306884765625,
0.0191192626953125,
-0.044403076171875,
0.032012939453125,
0.01285552978515625,
0.0450439453125,
0.00905609130859375,
-0.0211639404296875,
0.04302978515625,
-0.00019812583923339844,
-0.036590576171875,
-0.01009368896484375,
0.079345703125,
0.026519775390625,
0.0166473388671875,
0.0112152099609375,
-0.017578125,
-0.0282135009765625,
-0.0004038810729980469,
-0.09735107421875,
-0.0288543701171875,
0.0198822021484375,
-0.053680419921875,
-0.030364990234375,
0.0141448974609375,
-0.035980224609375,
-0.006175994873046875,
-0.00644683837890625,
0.049652099609375,
-0.0382080078125,
-0.0275115966796875,
-0.0008521080017089844,
-0.016571044921875,
0.0186767578125,
0.017242431640625,
-0.041717529296875,
0.01457977294921875,
0.0169525146484375,
0.0816650390625,
0.008331298828125,
-0.034942626953125,
-0.0171051025390625,
-0.02911376953125,
-0.0210418701171875,
0.0272064208984375,
-0.00392913818359375,
0.0006461143493652344,
-0.0244140625,
0.0298309326171875,
-0.0118865966796875,
-0.05389404296875,
0.0197906494140625,
-0.0194091796875,
0.01058197021484375,
-0.0016384124755859375,
-0.0243988037109375,
-0.041351318359375,
0.0233001708984375,
-0.040130615234375,
0.0892333984375,
0.02557373046875,
-0.06744384765625,
0.023712158203125,
-0.04534912109375,
-0.0096282958984375,
-0.01885986328125,
0.0033245086669921875,
-0.0789794921875,
-0.00588226318359375,
0.00914764404296875,
0.0618896484375,
-0.0231170654296875,
0.0050506591796875,
-0.0411376953125,
-0.0178680419921875,
0.0218658447265625,
-0.0016050338745117188,
0.080322265625,
0.0172271728515625,
-0.03814697265625,
0.024749755859375,
-0.05133056640625,
0.0170440673828125,
0.040985107421875,
-0.0200042724609375,
-0.0066070556640625,
-0.051605224609375,
0.011474609375,
0.0185394287109375,
0.0135955810546875,
-0.0421142578125,
0.0148468017578125,
-0.0133819580078125,
0.032806396484375,
0.047576904296875,
-0.01512908935546875,
0.027313232421875,
-0.026702880859375,
0.0201873779296875,
0.01959228515625,
0.017913818359375,
-0.00759124755859375,
-0.035003662109375,
-0.0655517578125,
-0.0386962890625,
0.0298309326171875,
0.028350830078125,
-0.036651611328125,
0.03680419921875,
-0.0130615234375,
-0.06268310546875,
-0.037353515625,
0.005584716796875,
0.0396728515625,
0.04364013671875,
0.024627685546875,
-0.032623291015625,
-0.035614013671875,
-0.07415771484375,
-0.0007815361022949219,
0.01027679443359375,
0.0037822723388671875,
0.0269012451171875,
0.05120849609375,
-0.00720977783203125,
0.038055419921875,
-0.031341552734375,
-0.016754150390625,
-0.023681640625,
0.006603240966796875,
0.0338134765625,
0.06396484375,
0.062469482421875,
-0.046844482421875,
-0.04345703125,
-0.0099029541015625,
-0.06939697265625,
0.011688232421875,
-0.007228851318359375,
-0.0180206298828125,
0.011016845703125,
0.0138092041015625,
-0.046234130859375,
0.041717529296875,
0.0197296142578125,
-0.025634765625,
0.03521728515625,
-0.020965576171875,
0.0179595947265625,
-0.07977294921875,
0.013885498046875,
0.03155517578125,
-0.0135040283203125,
-0.039276123046875,
0.0119171142578125,
0.00826263427734375,
-0.005352020263671875,
-0.037078857421875,
0.046112060546875,
-0.04498291015625,
-0.0115814208984375,
-0.00875091552734375,
-0.0232086181640625,
-0.00030303001403808594,
0.05401611328125,
-0.0126800537109375,
0.0261688232421875,
0.060821533203125,
-0.035400390625,
0.03277587890625,
0.02362060546875,
-0.0191650390625,
0.02520751953125,
-0.056884765625,
0.01183319091796875,
0.0046844482421875,
0.0176239013671875,
-0.072265625,
-0.0160980224609375,
0.02294921875,
-0.04217529296875,
0.049652099609375,
-0.033660888671875,
-0.031707763671875,
-0.03497314453125,
-0.035064697265625,
0.0269622802734375,
0.049072265625,
-0.054656982421875,
0.033966064453125,
0.0171356201171875,
0.0252685546875,
-0.0445556640625,
-0.064208984375,
-0.0225830078125,
-0.0298309326171875,
-0.0609130859375,
0.0311279296875,
0.0057525634765625,
0.005153656005859375,
0.01177215576171875,
0.00012010335922241211,
-0.008697509765625,
-0.0035533905029296875,
0.035125732421875,
0.023162841796875,
-0.024200439453125,
-0.0115966796875,
-0.0219879150390625,
-0.000797271728515625,
0.002704620361328125,
-0.0214691162109375,
0.04168701171875,
-0.020660400390625,
-0.006969451904296875,
-0.06378173828125,
-0.0031280517578125,
0.03582763671875,
-0.001064300537109375,
0.06683349609375,
0.08563232421875,
-0.039306640625,
-0.004199981689453125,
-0.031341552734375,
-0.0265960693359375,
-0.037353515625,
0.032440185546875,
-0.0300140380859375,
-0.033721923828125,
0.06463623046875,
-0.00629425048828125,
0.0119171142578125,
0.05224609375,
0.024993896484375,
-0.0069732666015625,
0.052093505859375,
0.04705810546875,
0.0178070068359375,
0.058319091796875,
-0.0816650390625,
-0.017578125,
-0.06304931640625,
-0.0298919677734375,
-0.031036376953125,
-0.05535888671875,
-0.046844482421875,
-0.0200042724609375,
0.035064697265625,
0.0170440673828125,
-0.0401611328125,
0.036651611328125,
-0.063720703125,
0.00579833984375,
0.05364990234375,
0.04052734375,
-0.030548095703125,
0.018829345703125,
-0.0218505859375,
-0.00020575523376464844,
-0.060821533203125,
-0.0135498046875,
0.08050537109375,
0.032684326171875,
0.04766845703125,
-0.01000213623046875,
0.04864501953125,
-0.0164337158203125,
0.0283966064453125,
-0.042938232421875,
0.04327392578125,
-0.01528167724609375,
-0.033294677734375,
-0.0167694091796875,
-0.038787841796875,
-0.08203125,
0.01329803466796875,
-0.022705078125,
-0.047698974609375,
0.0108795166015625,
0.01715087890625,
-0.0146636962890625,
0.05487060546875,
-0.0673828125,
0.0736083984375,
-0.01103973388671875,
-0.02978515625,
0.0010080337524414062,
-0.054534912109375,
0.0225830078125,
0.0188140869140625,
-0.0156402587890625,
-0.00592803955078125,
0.0118255615234375,
0.08184814453125,
-0.0517578125,
0.0645751953125,
-0.044189453125,
0.032012939453125,
0.036956787109375,
-0.00936126708984375,
0.028961181640625,
-0.0147247314453125,
-0.01320648193359375,
0.027862548828125,
-0.001323699951171875,
-0.034881591796875,
-0.040618896484375,
0.043060302734375,
-0.0743408203125,
-0.02227783203125,
-0.0157012939453125,
-0.032562255859375,
0.0185699462890625,
0.0123138427734375,
0.04473876953125,
0.0526123046875,
0.0141754150390625,
0.02606201171875,
0.0428466796875,
-0.03143310546875,
0.036346435546875,
-0.00650787353515625,
-0.0096435546875,
-0.03546142578125,
0.05853271484375,
0.024322509765625,
0.01306915283203125,
0.00815582275390625,
0.022186279296875,
-0.0203857421875,
-0.04534912109375,
-0.031524658203125,
0.0225067138671875,
-0.053802490234375,
-0.042724609375,
-0.05035400390625,
-0.035736083984375,
-0.028656005859375,
-0.00893402099609375,
-0.04034423828125,
-0.0291595458984375,
-0.027618408203125,
0.0158538818359375,
0.052825927734375,
0.0430908203125,
-0.0120697021484375,
0.047119140625,
-0.038970947265625,
0.00957489013671875,
0.00583648681640625,
0.03277587890625,
0.0048065185546875,
-0.0662841796875,
-0.024017333984375,
-0.006763458251953125,
-0.03411865234375,
-0.04998779296875,
0.03948974609375,
0.0199127197265625,
0.043060302734375,
0.02923583984375,
-0.0146484375,
0.05389404296875,
-0.00494384765625,
0.0418701171875,
0.03656005859375,
-0.04486083984375,
0.045135498046875,
0.00001996755599975586,
0.0161285400390625,
0.00707244873046875,
0.030731201171875,
-0.01422119140625,
-0.0030517578125,
-0.0758056640625,
-0.05889892578125,
0.062744140625,
0.00791168212890625,
0.0009813308715820312,
0.0284881591796875,
0.0557861328125,
0.0081329345703125,
0.0009813308715820312,
-0.0572509765625,
-0.03717041015625,
-0.0278167724609375,
-0.022430419921875,
0.002918243408203125,
-0.0034770965576171875,
-0.00624847412109375,
-0.055450439453125,
0.055023193359375,
-0.007061004638671875,
0.057769775390625,
0.022857666015625,
-0.007289886474609375,
-0.003948211669921875,
-0.033721923828125,
0.032379150390625,
0.02203369140625,
-0.0224609375,
0.00624847412109375,
0.0148468017578125,
-0.04351806640625,
0.0093231201171875,
0.0110015869140625,
0.00007593631744384766,
-0.0013942718505859375,
0.040252685546875,
0.07452392578125,
0.0033855438232421875,
0.0055999755859375,
0.032012939453125,
-0.00861358642578125,
-0.032196044921875,
-0.0171356201171875,
0.013031005859375,
0.00252532958984375,
0.031402587890625,
0.0186614990234375,
0.02996826171875,
-0.00890350341796875,
-0.0173187255859375,
0.021148681640625,
0.04217529296875,
-0.01898193359375,
-0.0274658203125,
0.047882080078125,
-0.00921630859375,
-0.01239013671875,
0.06756591796875,
-0.01019287109375,
-0.035003662109375,
0.08563232421875,
0.035552978515625,
0.0677490234375,
-0.00269317626953125,
-0.003086090087890625,
0.06585693359375,
0.01910400390625,
-0.0003581047058105469,
0.01335906982421875,
0.01239776611328125,
-0.06585693359375,
0.006755828857421875,
-0.04315185546875,
0.006366729736328125,
0.027130126953125,
-0.03961181640625,
0.024200439453125,
-0.052947998046875,
-0.0310821533203125,
0.0159912109375,
0.0295562744140625,
-0.07830810546875,
0.0194091796875,
-0.0081634521484375,
0.07135009765625,
-0.05303955078125,
0.06072998046875,
0.06475830078125,
-0.0369873046875,
-0.08758544921875,
-0.01522064208984375,
0.0006122589111328125,
-0.0703125,
0.0474853515625,
0.036468505859375,
0.01043701171875,
0.0066986083984375,
-0.06744384765625,
-0.04852294921875,
0.10821533203125,
0.04486083984375,
-0.010772705078125,
0.0221099853515625,
-0.0135650634765625,
0.0160369873046875,
-0.039947509765625,
0.0396728515625,
0.0151824951171875,
0.031768798828125,
0.0228729248046875,
-0.0499267578125,
0.0235595703125,
-0.0295562744140625,
0.00604248046875,
0.0121002197265625,
-0.06646728515625,
0.07080078125,
-0.036712646484375,
-0.0084686279296875,
-0.0003457069396972656,
0.0521240234375,
0.0121002197265625,
0.0111541748046875,
0.047515869140625,
0.06719970703125,
0.045013427734375,
-0.0194549560546875,
0.06927490234375,
-0.0011167526245117188,
0.04150390625,
0.042877197265625,
0.030975341796875,
0.034210205078125,
0.021514892578125,
-0.0175933837890625,
0.0266571044921875,
0.08319091796875,
-0.026336669921875,
0.0242919921875,
0.02093505859375,
0.005718231201171875,
-0.00579833984375,
0.009521484375,
-0.031341552734375,
0.040435791015625,
0.01427459716796875,
-0.043304443359375,
-0.01366424560546875,
0.0028076171875,
0.00298309326171875,
-0.0275726318359375,
-0.018829345703125,
0.039215087890625,
0.0029430389404296875,
-0.0287322998046875,
0.06805419921875,
0.00896453857421875,
0.072998046875,
-0.036224365234375,
0.00417327880859375,
-0.020965576171875,
0.0160675048828125,
-0.0258636474609375,
-0.054656982421875,
0.024261474609375,
-0.01953125,
-0.0007305145263671875,
-0.0003445148468017578,
0.05322265625,
-0.0265960693359375,
-0.035430908203125,
0.0192413330078125,
0.0240020751953125,
0.04229736328125,
0.004791259765625,
-0.09869384765625,
0.0141448974609375,
0.00559234619140625,
-0.054931640625,
0.027099609375,
0.03717041015625,
0.009613037109375,
0.05584716796875,
0.044952392578125,
-0.004032135009765625,
0.01036834716796875,
-0.00994873046875,
0.06390380859375,
-0.0361328125,
-0.0179595947265625,
-0.0592041015625,
0.04827880859375,
-0.0113983154296875,
-0.045379638671875,
0.0341796875,
0.03924560546875,
0.06268310546875,
-0.00286865234375,
0.0297393798828125,
-0.0245361328125,
-0.00742340087890625,
-0.0300140380859375,
0.061798095703125,
-0.05926513671875,
-0.0038509368896484375,
-0.0122528076171875,
-0.053802490234375,
-0.02392578125,
0.05474853515625,
-0.008697509765625,
0.033111572265625,
0.0364990234375,
0.07379150390625,
-0.0275115966796875,
-0.02191162109375,
0.0176239013671875,
0.010162353515625,
0.00730133056640625,
0.0259246826171875,
0.0264739990234375,
-0.055450439453125,
0.0244598388671875,
-0.05328369140625,
-0.021270751953125,
-0.009124755859375,
-0.05596923828125,
-0.06610107421875,
-0.06756591796875,
-0.043914794921875,
-0.0467529296875,
-0.0166473388671875,
0.07275390625,
0.0850830078125,
-0.049072265625,
-0.012420654296875,
0.003662109375,
0.01027679443359375,
-0.024444580078125,
-0.0171356201171875,
0.049560546875,
-0.0198211669921875,
-0.054534912109375,
-0.024810791015625,
-0.004688262939453125,
0.0274658203125,
-0.0005803108215332031,
-0.01445770263671875,
-0.01378631591796875,
-0.0181121826171875,
0.0196990966796875,
0.01352691650390625,
-0.0418701171875,
-0.01397705078125,
-0.0171661376953125,
-0.01023101806640625,
0.0239715576171875,
0.0380859375,
-0.033538818359375,
0.0171966552734375,
0.0283966064453125,
0.031036376953125,
0.061981201171875,
-0.0283966064453125,
0.006805419921875,
-0.06329345703125,
0.041748046875,
-0.01120758056640625,
0.0379638671875,
0.032562255859375,
-0.031036376953125,
0.047088623046875,
0.0290374755859375,
-0.036041259765625,
-0.06304931640625,
-0.0110015869140625,
-0.0791015625,
-0.0138092041015625,
0.06451416015625,
-0.03851318359375,
-0.0406494140625,
0.0404052734375,
0.004390716552734375,
0.04937744140625,
-0.00824737548828125,
0.0312347412109375,
0.020751953125,
-0.014617919921875,
-0.04669189453125,
-0.037353515625,
0.034393310546875,
0.012542724609375,
-0.04571533203125,
-0.0311279296875,
-0.0013799667358398438,
0.050933837890625,
0.0144805908203125,
0.03790283203125,
-0.00780487060546875,
0.0051116943359375,
0.01153564453125,
0.0418701171875,
-0.031707763671875,
-0.00667572021484375,
-0.020843505859375,
0.00524139404296875,
-0.00849151611328125,
-0.04449462890625
]
] |
jsylee/scibert_scivocab_uncased-finetuned-ner | 2021-11-22T03:52:41.000Z | [
"transformers",
"pytorch",
"bert",
"token-classification",
"Named Entity Recognition",
"SciBERT",
"Adverse Effect",
"Drug",
"Medical",
"en",
"dataset:ade_corpus_v2",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | token-classification | jsylee | null | null | jsylee/scibert_scivocab_uncased-finetuned-ner | 11 | 7,397 | transformers | 2022-03-02T23:29:05 | ---
language:
- en
tags:
- Named Entity Recognition
- SciBERT
- Adverse Effect
- Drug
- Medical
datasets:
- ade_corpus_v2
widget:
- text: "Abortion, miscarriage or uterine hemorrhage associated with misoprostol (Cytotec), a labor-inducing drug."
example_title: "Abortion, miscarriage, ..."
- text: "Addiction to many sedatives and analgesics, such as diazepam, morphine, etc."
example_title: "Addiction to many..."
- text: "Birth defects associated with thalidomide"
example_title: "Birth defects associated..."
- text: "Bleeding of the intestine associated with aspirin therapy"
example_title: "Bleeding of the intestine..."
- text: "Cardiovascular disease associated with COX-2 inhibitors (i.e. Vioxx)"
example_title: "Cardiovascular disease..."
---
This is a SciBERT-based model fine-tuned to perform Named Entity Recognition for drug names and adverse drug effects.
This model classifies input tokens into one of five classes:
- `B-DRUG`: beginning of a drug entity
- `I-DRUG`: within a drug entity
- `B-EFFECT`: beginning of an AE entity
- `I-EFFECT`: within an AE entity
- `O`: outside either of the above entities
To get started using this model for inference, simply set up an NER `pipeline` like below:
```python
from transformers import (AutoModelForTokenClassification,
AutoTokenizer,
pipeline,
)
model_checkpoint = "jsylee/scibert_scivocab_uncased-finetuned-ner"
model = AutoModelForTokenClassification.from_pretrained(model_checkpoint, num_labels=5,
id2label={0: 'O', 1: 'B-DRUG', 2: 'I-DRUG', 3: 'B-EFFECT', 4: 'I-EFFECT'}
)
tokenizer = AutoTokenizer.from_pretrained(model_checkpoint)
model_pipeline = pipeline(task="ner", model=model, tokenizer=tokenizer)
print( model_pipeline ("Abortion, miscarriage or uterine hemorrhage associated with misoprostol (Cytotec), a labor-inducing drug."))
```
SciBERT: https://huggingface.co/allenai/scibert_scivocab_uncased
Dataset: https://huggingface.co/datasets/ade_corpus_v2
| 2,333 | [
[
-0.0011806488037109375,
-0.0213165283203125,
0.03411865234375,
0.01190948486328125,
-0.0084991455078125,
-0.0033626556396484375,
0.01146697998046875,
-0.031158447265625,
0.035003662109375,
0.04217529296875,
-0.028900146484375,
-0.053558349609375,
-0.059539794921875,
0.01502227783203125,
-0.03912353515625,
0.10626220703125,
0.004611968994140625,
0.036102294921875,
0.0145721435546875,
-0.0102996826171875,
-0.0173492431640625,
-0.0231781005859375,
-0.048980712890625,
-0.032989501953125,
0.0577392578125,
0.0008821487426757812,
0.032989501953125,
0.035552978515625,
0.04193115234375,
0.0181732177734375,
-0.01605224609375,
-0.0129852294921875,
-0.01561737060546875,
0.006587982177734375,
-0.02130126953125,
-0.02191162109375,
-0.038177490234375,
-0.0006961822509765625,
0.031890869140625,
0.033477783203125,
-0.0018825531005859375,
0.01727294921875,
-0.00807952880859375,
0.0253143310546875,
-0.06805419921875,
-0.0013875961303710938,
-0.034515380859375,
0.0162353515625,
-0.0018291473388671875,
0.00798797607421875,
-0.032012939453125,
-0.006114959716796875,
0.0211029052734375,
-0.06048583984375,
0.0205535888671875,
0.0226287841796875,
0.09649658203125,
0.038665771484375,
-0.035552978515625,
-0.04071044921875,
-0.041900634765625,
0.040985107421875,
-0.050689697265625,
0.0269927978515625,
0.0016107559204101562,
0.0000807046890258789,
-0.0264434814453125,
-0.08172607421875,
-0.035736083984375,
-0.0088958740234375,
-0.032562255859375,
0.006427764892578125,
-0.0219268798828125,
0.026580810546875,
0.036834716796875,
0.0222930908203125,
-0.046051025390625,
0.01300811767578125,
-0.06915283203125,
-0.050811767578125,
0.0277099609375,
0.0269927978515625,
0.021240234375,
-0.03192138671875,
-0.0489501953125,
0.02203369140625,
-0.0438232421875,
0.003833770751953125,
0.00458526611328125,
0.01412200927734375,
-0.0031757354736328125,
0.05133056640625,
-0.011871337890625,
0.06170654296875,
0.0237579345703125,
0.01515960693359375,
0.034576416015625,
-0.0070953369140625,
-0.04095458984375,
0.02813720703125,
0.0828857421875,
0.0242767333984375,
-0.00902557373046875,
0.001964569091796875,
-0.003009796142578125,
0.0022487640380859375,
0.01154327392578125,
-0.0985107421875,
-0.064208984375,
0.04052734375,
-0.042327880859375,
-0.04046630859375,
-0.01047515869140625,
-0.04339599609375,
-0.0477294921875,
-0.015655517578125,
0.06085205078125,
-0.04180908203125,
0.016326904296875,
0.004070281982421875,
-0.0160064697265625,
0.0140838623046875,
0.0109710693359375,
-0.050018310546875,
0.0091400146484375,
0.022003173828125,
0.068603515625,
-0.007144927978515625,
-0.017913818359375,
-0.05072021484375,
0.003154754638671875,
-0.01788330078125,
0.0721435546875,
-0.0248565673828125,
-0.020538330078125,
-0.039581298828125,
0.0224609375,
-0.0173187255859375,
-0.0303802490234375,
0.041290283203125,
-0.0207672119140625,
0.02734375,
-0.0013799667358398438,
-0.05712890625,
-0.0118255615234375,
-0.0013875961303710938,
-0.043731689453125,
0.0386962890625,
0.006938934326171875,
-0.080810546875,
0.052154541015625,
-0.058929443359375,
-0.0152740478515625,
0.0162811279296875,
-0.0125274658203125,
-0.04852294921875,
-0.00556182861328125,
0.0033054351806640625,
0.03662109375,
-0.007602691650390625,
0.0234375,
-0.027587890625,
-0.039306640625,
0.0169219970703125,
0.0009889602661132812,
0.06951904296875,
0.0305633544921875,
-0.00565338134765625,
0.0304412841796875,
-0.06396484375,
0.0017986297607421875,
-0.01007843017578125,
-0.018798828125,
-0.024169921875,
-0.024749755859375,
0.028106689453125,
0.02783203125,
0.008941650390625,
-0.04815673828125,
0.019439697265625,
-0.0272674560546875,
0.044708251953125,
0.033172607421875,
0.0254058837890625,
0.021484375,
0.0018014907836914062,
0.0224609375,
0.019927978515625,
0.01070404052734375,
0.003330230712890625,
-0.0230560302734375,
-0.0667724609375,
-0.054107666015625,
0.06005859375,
0.045684814453125,
-0.03143310546875,
0.032073974609375,
0.011993408203125,
-0.034210205078125,
-0.03759765625,
-0.018798828125,
0.0233154296875,
0.054962158203125,
0.038818359375,
-0.0283203125,
-0.057037353515625,
-0.076416015625,
-0.0191802978515625,
-0.0019989013671875,
-0.014129638671875,
0.011688232421875,
0.06512451171875,
-0.040771484375,
0.058074951171875,
-0.0494384765625,
-0.035003662109375,
0.0082855224609375,
0.013397216796875,
0.044586181640625,
0.04254150390625,
0.0271148681640625,
-0.03680419921875,
-0.016510009765625,
-0.0297698974609375,
-0.046722412109375,
-0.01245880126953125,
-0.01544952392578125,
-0.0230255126953125,
-0.01194000244140625,
0.039794921875,
-0.045440673828125,
0.045318603515625,
0.014434814453125,
-0.04364013671875,
0.0418701171875,
-0.0289306640625,
0.002902984619140625,
-0.09039306640625,
0.0233612060546875,
0.0028705596923828125,
0.0004763603210449219,
-0.0677490234375,
-0.0136871337890625,
0.0254058837890625,
-0.00945281982421875,
-0.060089111328125,
0.046966552734375,
-0.0220794677734375,
0.01763916015625,
-0.01151275634765625,
-0.02044677734375,
0.021148681640625,
0.0185699462890625,
0.0198211669921875,
0.037933349609375,
0.061614990234375,
-0.05426025390625,
0.016998291015625,
0.0347900390625,
-0.0046539306640625,
0.0458984375,
-0.0733642578125,
-0.020660400390625,
-0.0107574462890625,
0.01800537109375,
-0.03082275390625,
-0.038848876953125,
0.055206298828125,
-0.032867431640625,
0.032135009765625,
-0.04486083984375,
-0.0219268798828125,
-0.0308990478515625,
0.003673553466796875,
0.0136871337890625,
0.01181793212890625,
-0.037384033203125,
0.0670166015625,
0.0233154296875,
0.0128631591796875,
-0.060211181640625,
-0.04705810546875,
-0.0036373138427734375,
-0.040130615234375,
-0.00726318359375,
0.03692626953125,
-0.001323699951171875,
0.0032825469970703125,
-0.0010824203491210938,
-0.0032329559326171875,
-0.0263519287109375,
-0.005828857421875,
0.046966552734375,
0.046722412109375,
0.01464080810546875,
0.0308685302734375,
0.0265350341796875,
-0.0198211669921875,
0.01297760009765625,
-0.0139617919921875,
0.0311279296875,
-0.015411376953125,
0.00240325927734375,
-0.029205322265625,
0.015380859375,
0.04351806640625,
-0.019073486328125,
0.048675537109375,
0.055999755859375,
-0.0582275390625,
-0.0082855224609375,
-0.01206207275390625,
-0.022613525390625,
-0.032440185546875,
0.028167724609375,
-0.02130126953125,
-0.034881591796875,
0.051727294921875,
-0.002071380615234375,
-0.01000213623046875,
0.045654296875,
0.045867919921875,
-0.01116180419921875,
0.06402587890625,
0.0227508544921875,
0.0036563873291015625,
0.0131378173828125,
-0.035614013671875,
0.032745361328125,
-0.0765380859375,
-0.026458740234375,
-0.023223876953125,
-0.0172576904296875,
-0.0479736328125,
-0.0275115966796875,
0.0274810791015625,
0.0224609375,
-0.04107666015625,
0.053802490234375,
-0.0667724609375,
0.0263671875,
0.03521728515625,
0.005138397216796875,
0.01154327392578125,
-0.0186004638671875,
-0.0264434814453125,
0.0017919540405273438,
-0.047210693359375,
-0.031005859375,
0.0772705078125,
0.0221099853515625,
0.061065673828125,
-0.02001953125,
0.07354736328125,
0.001056671142578125,
0.0386962890625,
-0.0631103515625,
0.0256195068359375,
-0.013916015625,
-0.062164306640625,
0.0160064697265625,
-0.0228729248046875,
-0.07275390625,
0.0223846435546875,
-0.050537109375,
-0.0694580078125,
0.05072021484375,
0.0218048095703125,
-0.0579833984375,
0.01434326171875,
-0.05029296875,
0.07403564453125,
-0.0201568603515625,
-0.0294952392578125,
-0.0077667236328125,
-0.046630859375,
0.01543426513671875,
-0.00536346435546875,
-0.0011281967163085938,
-0.020538330078125,
0.01174163818359375,
0.057647705078125,
-0.03289794921875,
0.047210693359375,
-0.0128326416015625,
0.025054931640625,
0.007080078125,
-0.0011348724365234375,
0.0125274658203125,
0.0186614990234375,
0.0017347335815429688,
0.03350830078125,
0.019683837890625,
-0.0109405517578125,
-0.0289459228515625,
0.062164306640625,
-0.04119873046875,
-0.002716064453125,
-0.058837890625,
-0.0401611328125,
0.031951904296875,
0.04119873046875,
0.0545654296875,
0.0404052734375,
0.0027637481689453125,
-0.0103302001953125,
0.03472900390625,
-0.035980224609375,
0.018463134765625,
0.065673828125,
-0.01470947265625,
-0.033203125,
0.0548095703125,
0.0133056640625,
-0.00873565673828125,
0.0083770751953125,
0.00027680397033691406,
-0.020782470703125,
-0.032318115234375,
-0.04156494140625,
0.01554107666015625,
-0.044830322265625,
-0.03106689453125,
-0.07293701171875,
-0.0262298583984375,
-0.028656005859375,
0.0058441162109375,
-0.033203125,
-0.0230865478515625,
-0.047393798828125,
-0.0004069805145263672,
0.034881591796875,
0.0455322265625,
-0.0167388916015625,
0.010498046875,
-0.042388916015625,
0.01032257080078125,
-0.007419586181640625,
0.026824951171875,
-0.0377197265625,
-0.041168212890625,
-0.00018596649169921875,
-0.0125579833984375,
-0.02789306640625,
-0.07757568359375,
0.049713134765625,
0.01494598388671875,
0.02978515625,
0.0299224853515625,
0.01397705078125,
0.05145263671875,
-0.033172607421875,
0.0313720703125,
0.01311492919921875,
-0.0712890625,
0.060546875,
-0.0096588134765625,
-0.00534820556640625,
0.038818359375,
0.0433349609375,
-0.039093017578125,
-0.02423095703125,
-0.044769287109375,
-0.09478759765625,
0.0428466796875,
0.0278472900390625,
-0.016998291015625,
-0.009613037109375,
0.036590576171875,
-0.0059967041015625,
0.01099395751953125,
-0.06243896484375,
-0.0330810546875,
-0.0042877197265625,
-0.0157318115234375,
0.03314208984375,
0.0029754638671875,
-0.005138397216796875,
-0.02166748046875,
0.06439208984375,
-0.00666046142578125,
0.027191162109375,
0.0369873046875,
-0.0227203369140625,
0.00980377197265625,
0.023651123046875,
0.027801513671875,
0.0285797119140625,
-0.03155517578125,
-0.001556396484375,
0.0213165283203125,
-0.0357666015625,
-0.0052642822265625,
0.046295166015625,
-0.0213165283203125,
-0.00604248046875,
0.0211944580078125,
0.0638427734375,
0.00122833251953125,
-0.04876708984375,
0.025665283203125,
-0.0182952880859375,
-0.008636474609375,
-0.0440673828125,
0.0241241455078125,
-0.015350341796875,
0.033843994140625,
0.019012451171875,
0.0267791748046875,
0.047210693359375,
-0.01971435546875,
0.031158447265625,
0.0245361328125,
-0.0498046875,
-0.0005216598510742188,
0.0693359375,
-0.00948333740234375,
-0.019073486328125,
0.04730224609375,
-0.0211029052734375,
-0.043975830078125,
0.051910400390625,
0.023956298828125,
0.047943115234375,
-0.005039215087890625,
0.0081024169921875,
0.062042236328125,
-0.004505157470703125,
0.01171875,
0.01509857177734375,
0.01490020751953125,
-0.04730224609375,
-0.01340484619140625,
-0.055633544921875,
-0.0262603759765625,
0.01447296142578125,
-0.061065673828125,
0.05059814453125,
-0.033050537109375,
-0.0294647216796875,
0.03692626953125,
-0.01259613037109375,
-0.052276611328125,
0.0293731689453125,
-0.0013933181762695312,
0.041900634765625,
-0.08209228515625,
0.0584716796875,
0.05926513671875,
-0.034088134765625,
-0.07733154296875,
-0.01120758056640625,
-0.00850677490234375,
-0.036956787109375,
0.056854248046875,
0.0267181396484375,
0.01197052001953125,
-0.005001068115234375,
-0.0239410400390625,
-0.07147216796875,
0.09735107421875,
0.00501251220703125,
-0.059539794921875,
-0.0097198486328125,
-0.016998291015625,
0.059661865234375,
-0.03363037109375,
0.045623779296875,
0.03912353515625,
0.01505279541015625,
0.008026123046875,
-0.054534912109375,
0.0033779144287109375,
-0.031463623046875,
-0.0177001953125,
0.018829345703125,
-0.0182037353515625,
0.06903076171875,
-0.05517578125,
-0.005828857421875,
0.01029205322265625,
0.0323486328125,
0.040313720703125,
0.04022216796875,
0.0369873046875,
0.0404052734375,
0.0560302734375,
-0.014434814453125,
0.05633544921875,
-0.048614501953125,
0.062164306640625,
0.08709716796875,
-0.00975799560546875,
0.042083740234375,
0.040740966796875,
0.0146942138671875,
0.04901123046875,
0.05633544921875,
-0.042755126953125,
0.045074462890625,
0.030609130859375,
-0.006076812744140625,
-0.01163482666015625,
-0.00865936279296875,
-0.019012451171875,
0.0345458984375,
0.052947998046875,
-0.06536865234375,
-0.00644683837890625,
0.01776123046875,
-0.00034499168395996094,
-0.0400390625,
-0.02203369140625,
0.03533935546875,
-0.00286865234375,
-0.036468505859375,
0.032470703125,
-0.001049041748046875,
0.043609619140625,
-0.0231781005859375,
-0.0124053955078125,
0.0140533447265625,
0.033355712890625,
-0.029693603515625,
-0.045074462890625,
0.0168914794921875,
-0.0106353759765625,
-0.01177215576171875,
0.02978515625,
0.061126708984375,
-0.040008544921875,
-0.0491943359375,
0.028564453125,
0.00342559814453125,
0.0183563232421875,
0.0182342529296875,
-0.067138671875,
-0.006259918212890625,
-0.0103607177734375,
-0.00609588623046875,
0.004322052001953125,
-0.0034942626953125,
-0.007511138916015625,
0.053863525390625,
0.039459228515625,
0.0217132568359375,
-0.0154571533203125,
0.00605010986328125,
0.0623779296875,
-0.03564453125,
-0.025238037109375,
-0.0479736328125,
0.05029296875,
-0.010955810546875,
-0.04119873046875,
0.031341552734375,
0.059051513671875,
0.033477783203125,
-0.01088714599609375,
0.0301971435546875,
-0.0261383056640625,
0.032989501953125,
-0.022979736328125,
0.07733154296875,
-0.024810791015625,
0.0192108154296875,
-0.0291595458984375,
-0.0418701171875,
-0.027374267578125,
0.076416015625,
-0.0260162353515625,
-0.009918212890625,
0.05010986328125,
0.07147216796875,
-0.0079345703125,
-0.0100860595703125,
-0.0032215118408203125,
0.035980224609375,
0.0189361572265625,
0.04840087890625,
0.041717529296875,
-0.054443359375,
0.0187225341796875,
-0.030792236328125,
-0.0084228515625,
-0.002880096435546875,
-0.059844970703125,
-0.075439453125,
-0.031524658203125,
-0.041351318359375,
-0.0614013671875,
0.00904083251953125,
0.09088134765625,
0.06732177734375,
-0.0838623046875,
0.0260772705078125,
-0.01329803466796875,
-0.0026607513427734375,
0.005443572998046875,
-0.0203704833984375,
0.03717041015625,
0.004589080810546875,
-0.052947998046875,
0.012237548828125,
0.01157379150390625,
0.009246826171875,
0.002445220947265625,
0.0010623931884765625,
-0.002704620361328125,
-0.0005145072937011719,
0.0255279541015625,
0.01422882080078125,
-0.04296875,
-0.017364501953125,
0.000652313232421875,
-0.006328582763671875,
-0.01019287109375,
0.058502197265625,
-0.0679931640625,
0.037872314453125,
0.0286712646484375,
-0.00203704833984375,
0.041259765625,
-0.025146484375,
0.02825927734375,
-0.0251617431640625,
0.00293731689453125,
0.0296630859375,
0.05474853515625,
0.01494598388671875,
-0.04962158203125,
0.032562255859375,
0.0095062255859375,
-0.053131103515625,
-0.06292724609375,
0.0063629150390625,
-0.058319091796875,
-0.01520538330078125,
0.059326171875,
-0.01119232177734375,
-0.008056640625,
-0.01091766357421875,
-0.015533447265625,
0.0322265625,
-0.038421630859375,
0.056549072265625,
0.01910400390625,
-0.03631591796875,
-0.0001647472381591797,
-0.0352783203125,
0.0272369384765625,
0.01428985595703125,
-0.045074462890625,
-0.01428985595703125,
0.038818359375,
0.0625,
0.025421142578125,
0.035858154296875,
-0.01509857177734375,
0.02215576171875,
-0.01329803466796875,
0.0210418701171875,
-0.00705718994140625,
-0.0233917236328125,
-0.0299224853515625,
-0.00972747802734375,
-0.018890380859375,
-0.0179443359375
]
] |
nomic-ai/gpt4all-j | 2023-06-02T22:29:39.000Z | [
"transformers",
"pytorch",
"gptj",
"text-generation",
"en",
"dataset:nomic-ai/gpt4all-j-prompt-generations",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | text-generation | nomic-ai | null | null | nomic-ai/gpt4all-j | 257 | 7,395 | transformers | 2023-04-11T15:39:16 | ---
license: apache-2.0
datasets:
- nomic-ai/gpt4all-j-prompt-generations
language:
- en
pipeline_tag: text-generation
---
# Model Card for GPT4All-J
An Apache-2 licensed chatbot trained over a massive curated corpus of assistant interactions including word problems, multi-turn dialogue, code, poems, songs, and stories.
## Model Details
### Model Description
<!-- Provide a longer summary of what this model is. -->
This model has been finetuned from [GPT-J](https://huggingface.co/EleutherAI/gpt-j-6B)
- **Developed by:** [Nomic AI](https://home.nomic.ai)
- **Model Type:** A finetuned GPT-J model on assistant style interaction data
- **Language(s) (NLP):** English
- **License:** Apache-2
- **Finetuned from model [optional]:** [GPT-J](https://huggingface.co/EleutherAI/gpt-j-6B)
We have released several versions of our finetuned GPT-J model using [different dataset versions](https://huggingface.co/datasets/nomic-ai/gpt4all-j-prompt-generations)
- v1.0: The original model trained on the v1.0 dataset
- v1.1-breezy: Trained on afiltered dataset where we removed all instances of AI language model
- v1.2-jazzy: Trained on a filtered dataset where we also removed instances like I'm sorry, I can't answer... and AI language model
- v1.3-groovy: We added Dolly and ShareGPT to the v1.2 dataset and removed ~8% of the dataset in v1.2 that contained semantic duplicates using [Atlas](https://atlas.nomic.ai/).
To download a model with a specific revision run
```python
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("nomic-ai/gpt4all-j", revision="v1.2-jazzy")
```
Downloading without specifying `revision` defaults to `main`/`v1.0`.
### Model Sources [optional]
<!-- Provide the basic links for the model. -->
- **Repository:** [https://github.com/nomic-ai/gpt4all](https://github.com/nomic-ai/gpt4all)
- **Base Model Repository:** [https://github.com/kingoflolz/mesh-transformer-jax](https://github.com/kingoflolz/mesh-transformer-jax)
- **Paper [optional]:** [GPT4All-J: An Apache-2 Licensed Assistant-Style Chatbot](https://s3.amazonaws.com/static.nomic.ai/gpt4all/2023_GPT4All-J_Technical_Report_2.pdf)
- **Demo [optional]:** [https://gpt4all.io/](https://gpt4all.io/)
### Training Procedure
GPT4All is made possible by our compute partner [Paperspace](https://www.paperspace.com/).
Trained on a DGX cluster with 8 A100 80GB GPUs for ~12 hours. Using Deepspeed + Accelerate, we use a global batch size of 256 with a learning rate of 2e-5. More information can be found in the repo.
### Results
Results on common sense reasoning benchmarks
```
| Model | BoolQ | PIQA | HellaSwag | WinoGrande | ARC-e | ARC-c | OBQA | Avg. |
|:--------------------------|:--------:|:--------:|:---------:|:----------:|:--------:|:--------:|:--------:|:--------:|
| GPT4All-J 6B v1.0 | 73.4 | 74.8 | 63.4 | 64.7 | 54.9 | 36.0 | 40.2 | 58.2 |
| GPT4All-J v1.1-breezy | 74.0 | 75.1 | 63.2 | 63.6 | 55.4 | 34.9 | 38.4 | 57.8 |
| GPT4All-J v1.2-jazzy | 74.8 | 74.9 | 63.6 | 63.8 | 56.6 | 35.3 | 41.0 | 58.6 |
| GPT4All-J v1.3-groovy | 73.6 | 74.3 | 63.8 | 63.5 | 57.7 | 35.0 | 38.8 | 58.1 |
| GPT4All-J Lora 6B | 68.6 | 75.8 | 66.2 | 63.5 | 56.4 | 35.7 | 40.2 | 58.1 |
| GPT4All LLaMa Lora 7B | 73.1 | 77.6 | 72.1 | 67.8 | 51.1 | 40.4 | 40.2 | 60.3 |
| GPT4All 13B snoozy | **83.3** | 79.2 | 75.0 | **71.3** | 60.9 | 44.2 | 43.4 | **65.3** |
| Dolly 6B | 68.8 | 77.3 | 67.6 | 63.9 | 62.9 | 38.7 | 41.2 | 60.1 |
| Dolly 12B | 56.7 | 75.4 | 71.0 | 62.2 | 64.6 | 38.5 | 40.4 | 58.4 |
| Alpaca 7B | 73.9 | 77.2 | 73.9 | 66.1 | 59.8 | 43.3 | 43.4 | 62.4 |
| Alpaca Lora 7B | 74.3 | **79.3** | 74.0 | 68.8 | 56.6 | 43.9 | 42.6 | 62.8 |
| GPT-J 6.7B | 65.4 | 76.2 | 66.2 | 64.1 | 62.2 | 36.6 | 38.2 | 58.4 |
| LLama 7B | 73.1 | 77.4 | 73.0 | 66.9 | 52.5 | 41.4 | 42.4 | 61.0 |
| LLama 13B | 68.5 | 79.1 | 76.2 | 70.1 | 60.0 | **44.6** | 42.2 | 63.0 |
| Pythia 6.7B | 63.5 | 76.3 | 64.0 | 61.1 | 61.3 | 35.2 | 37.2 | 57.0 |
| Pythia 12B | 67.7 | 76.6 | 67.3 | 63.8 | 63.9 | 34.8 | 38 | 58.9 |
| Fastchat T5 | 81.5 | 64.6 | 46.3 | 61.8 | 49.3 | 33.3 | 39.4 | 53.7 |
| Fastchat Vicuña 7B | 76.6 | 77.2 | 70.7 | 67.3 | 53.5 | 41.2 | 40.8 | 61.0 |
| Fastchat Vicuña 13B | 81.5 | 76.8 | 73.3 | 66.7 | 57.4 | 42.7 | 43.6 | 63.1 |
| StableVicuña RLHF | 82.3 | 78.6 | 74.1 | 70.9 | 61.0 | 43.5 | **44.4** | 65.0 |
| StableLM Tuned | 62.5 | 71.2 | 53.6 | 54.8 | 52.4 | 31.1 | 33.4 | 51.3 |
| StableLM Base | 60.1 | 67.4 | 41.2 | 50.1 | 44.9 | 27.0 | 32.0 | 42.2 |
| Koala 13B | 76.5 | 77.9 | 72.6 | 68.8 | 54.3 | 41.0 | 42.8 | 62.0 |
| Open Assistant Pythia 12B | 67.9 | 78.0 | 68.1 | 65.0 | 64.2 | 40.4 | 43.2 | 61.0 |
| Mosaic mpt-7b | 74.8 | **79.3** | **76.3** | 68.6 | **70.0** | 42.2 | 42.6 | 64.8 |
| text-davinci-003 | 88.1 | 83.8 | 83.4 | 75.8 | 83.9 | 63.9 | 51.0 | 75.7 |
``` | 6,013 | [
[
-0.0416259765625,
-0.043121337890625,
0.0229644775390625,
0.01071929931640625,
0.0036792755126953125,
0.01120758056640625,
0.0083770751953125,
-0.02496337890625,
0.0208892822265625,
0.0176849365234375,
-0.040435791015625,
-0.038177490234375,
-0.041015625,
-0.0013799667358398438,
0.0021877288818359375,
0.05682373046875,
0.0014324188232421875,
-0.017181396484375,
0.0008420944213867188,
-0.015655517578125,
-0.027862548828125,
-0.00943756103515625,
-0.044891357421875,
-0.0203704833984375,
0.01461029052734375,
0.0292816162109375,
0.0665283203125,
0.033294677734375,
0.0257568359375,
0.028717041015625,
-0.00905609130859375,
0.0101165771484375,
-0.0099029541015625,
-0.0270843505859375,
0.01505279541015625,
-0.0194549560546875,
-0.0369873046875,
0.0106201171875,
0.023406982421875,
0.034423828125,
-0.007171630859375,
0.0213775634765625,
0.00020694732666015625,
0.06536865234375,
-0.031585693359375,
0.0247039794921875,
-0.0247039794921875,
-0.00373077392578125,
-0.0024776458740234375,
-0.00461578369140625,
-0.003887176513671875,
-0.03271484375,
0.0103607177734375,
-0.05792236328125,
0.0125579833984375,
-0.003932952880859375,
0.1021728515625,
0.0257415771484375,
-0.0306854248046875,
-0.0174102783203125,
-0.027435302734375,
0.058929443359375,
-0.054718017578125,
0.02313232421875,
0.037109375,
0.011810302734375,
-0.0133819580078125,
-0.050201416015625,
-0.06524658203125,
0.0095367431640625,
-0.0258331298828125,
0.0384521484375,
-0.0172119140625,
-0.0199127197265625,
0.024200439453125,
0.036224365234375,
-0.055084228515625,
-0.01468658447265625,
-0.03466796875,
-0.00972747802734375,
0.054229736328125,
0.006931304931640625,
0.028411865234375,
-0.034149169921875,
-0.028228759765625,
-0.00732421875,
-0.0252227783203125,
0.030792236328125,
0.02947998046875,
-0.002651214599609375,
-0.03216552734375,
0.026611328125,
-0.0158233642578125,
0.040924072265625,
0.00609588623046875,
-0.023529052734375,
0.064697265625,
-0.033721923828125,
-0.0186004638671875,
-0.01311492919921875,
0.06451416015625,
0.031158447265625,
-0.0034351348876953125,
0.0080108642578125,
0.005252838134765625,
0.00911712646484375,
-0.0019702911376953125,
-0.040313720703125,
-0.041229248046875,
0.03460693359375,
-0.0230560302734375,
-0.01153564453125,
-0.00730133056640625,
-0.0531005859375,
0.0011434555053710938,
-0.0286865234375,
0.035980224609375,
-0.03411865234375,
-0.03857421875,
0.00616455078125,
-0.0218963623046875,
0.034210205078125,
0.0229034423828125,
-0.0745849609375,
0.019317626953125,
0.025909423828125,
0.0662841796875,
0.00225067138671875,
-0.014434814453125,
0.0107879638671875,
0.01416015625,
-0.032562255859375,
0.05029296875,
-0.0201263427734375,
-0.029449462890625,
-0.02703857421875,
0.0170135498046875,
-0.0209503173828125,
-0.01169586181640625,
0.04534912109375,
-0.0201568603515625,
0.042083740234375,
-0.020843505859375,
-0.032867431640625,
-0.025390625,
0.035003662109375,
-0.04632568359375,
0.0904541015625,
0.023651123046875,
-0.062042236328125,
0.0260009765625,
-0.061370849609375,
-0.00705718994140625,
-0.00791168212890625,
-0.0107879638671875,
-0.057861328125,
-0.031036376953125,
0.03662109375,
0.0222320556640625,
-0.0228424072265625,
0.01235198974609375,
-0.00519561767578125,
-0.02288818359375,
-0.0207061767578125,
-0.0214385986328125,
0.0928955078125,
0.02947998046875,
-0.0419921875,
0.0158233642578125,
-0.0596923828125,
0.006450653076171875,
0.031707763671875,
-0.0212249755859375,
0.003902435302734375,
-0.0222015380859375,
-0.0179901123046875,
0.017425537109375,
0.026947021484375,
-0.0254058837890625,
0.01348876953125,
-0.02001953125,
0.0433349609375,
0.0428466796875,
0.003841400146484375,
0.0321044921875,
-0.04522705078125,
0.025146484375,
0.0126953125,
0.0294647216796875,
-0.0096282958984375,
-0.053009033203125,
-0.0555419921875,
-0.043914794921875,
-0.00054931640625,
0.049957275390625,
-0.04315185546875,
0.0433349609375,
-0.01413726806640625,
-0.052398681640625,
-0.0316162109375,
-0.0014200210571289062,
0.0222015380859375,
0.040374755859375,
0.038909912109375,
-0.0031986236572265625,
-0.02923583984375,
-0.063720703125,
-0.011566162109375,
-0.01111602783203125,
0.0141754150390625,
0.04754638671875,
0.055572509765625,
-0.0179443359375,
0.07159423828125,
-0.038787841796875,
-0.0279388427734375,
-0.0168914794921875,
-0.0037841796875,
0.045318603515625,
0.041900634765625,
0.05572509765625,
-0.057098388671875,
-0.055328369140625,
0.0023021697998046875,
-0.061767578125,
0.00724029541015625,
-0.0038394927978515625,
-0.006076812744140625,
0.0193023681640625,
0.0183868408203125,
-0.06011962890625,
0.057464599609375,
0.04962158203125,
-0.0386962890625,
0.04998779296875,
-0.0236663818359375,
0.0179443359375,
-0.08575439453125,
0.01123809814453125,
-0.007579803466796875,
-0.007965087890625,
-0.03778076171875,
-0.0130615234375,
0.0035190582275390625,
-0.007755279541015625,
-0.0281829833984375,
0.050933837890625,
-0.056243896484375,
-0.00403594970703125,
0.01023101806640625,
-0.00174713134765625,
0.008331298828125,
0.0543212890625,
-0.0006976127624511719,
0.07598876953125,
0.0513916015625,
-0.0435791015625,
0.0279693603515625,
0.0249786376953125,
-0.028533935546875,
0.0217437744140625,
-0.0439453125,
0.00251007080078125,
0.00482940673828125,
0.0215911865234375,
-0.08319091796875,
-0.0193023681640625,
0.033782958984375,
-0.04925537109375,
0.016204833984375,
-0.00995635986328125,
-0.0224609375,
-0.07037353515625,
-0.0290679931640625,
0.00762176513671875,
0.041351318359375,
-0.040924072265625,
0.038299560546875,
0.029296875,
0.003082275390625,
-0.04620361328125,
-0.038238525390625,
-0.005779266357421875,
-0.02008056640625,
-0.055267333984375,
0.031585693359375,
-0.006679534912109375,
-0.009552001953125,
0.004077911376953125,
-0.017822265625,
-0.0085906982421875,
0.016754150390625,
0.0157470703125,
0.028961181640625,
-0.01096343994140625,
-0.0156097412109375,
-0.01503753662109375,
-0.00998687744140625,
-0.0093841552734375,
-0.00165557861328125,
0.043487548828125,
-0.0275726318359375,
-0.0237579345703125,
-0.05810546875,
-0.0024394989013671875,
0.035369873046875,
-0.0211181640625,
0.06439208984375,
0.050201416015625,
-0.01465606689453125,
0.017333984375,
-0.0369873046875,
-0.0068511962890625,
-0.036865234375,
0.005859375,
-0.052520751953125,
-0.06085205078125,
0.048797607421875,
0.0018796920776367188,
0.0149383544921875,
0.053924560546875,
0.0360107421875,
-0.0018939971923828125,
0.07867431640625,
0.0198822021484375,
-0.01178741455078125,
0.0322265625,
-0.052581787109375,
0.021240234375,
-0.05987548828125,
-0.039276123046875,
-0.036529541015625,
-0.019256591796875,
-0.0433349609375,
-0.0195159912109375,
0.024261474609375,
0.0148468017578125,
-0.049530029296875,
0.0272064208984375,
-0.0689697265625,
0.03546142578125,
0.06011962890625,
0.01325225830078125,
0.0081787109375,
-0.011566162109375,
-0.018280029296875,
0.0009207725524902344,
-0.04547119140625,
-0.03369140625,
0.08251953125,
0.0162506103515625,
0.0286407470703125,
0.032012939453125,
0.05108642578125,
0.0113067626953125,
0.000026106834411621094,
-0.03955078125,
0.0217742919921875,
0.0170440673828125,
-0.060272216796875,
-0.0289306640625,
-0.028411865234375,
-0.0855712890625,
0.033294677734375,
-0.0142364501953125,
-0.06500244140625,
0.016082763671875,
0.00812530517578125,
-0.0295562744140625,
0.031646728515625,
-0.06988525390625,
0.07659912109375,
-0.0186004638671875,
-0.0305938720703125,
-0.0019140243530273438,
-0.05328369140625,
0.04107666015625,
0.01403045654296875,
0.0292510986328125,
-0.02166748046875,
0.0144500732421875,
0.04998779296875,
-0.0501708984375,
0.037353515625,
-0.0189208984375,
0.01763916015625,
0.03497314453125,
-0.0105743408203125,
0.03875732421875,
0.01139068603515625,
0.00347137451171875,
0.011962890625,
0.00637054443359375,
-0.048431396484375,
-0.0217742919921875,
0.071044921875,
-0.09381103515625,
-0.06915283203125,
-0.058746337890625,
-0.03533935546875,
-0.0008726119995117188,
0.028228759765625,
0.023834228515625,
0.0171051025390625,
0.0173492431640625,
0.0071563720703125,
0.0374755859375,
-0.017333984375,
0.052520751953125,
0.027923583984375,
-0.0054931640625,
-0.033355712890625,
0.06439208984375,
0.00240325927734375,
0.00860595703125,
0.006114959716796875,
0.02032470703125,
-0.0316162109375,
-0.0303497314453125,
-0.03253173828125,
0.0161895751953125,
-0.022125244140625,
-0.01306915283203125,
-0.05389404296875,
-0.0232086181640625,
-0.061126708984375,
-0.01751708984375,
-0.0201873779296875,
-0.029876708984375,
-0.022979736328125,
-0.00527191162109375,
0.045623779296875,
0.0538330078125,
0.00363922119140625,
0.0229644775390625,
-0.0482177734375,
0.0258026123046875,
0.031219482421875,
0.01123809814453125,
-0.01038360595703125,
-0.0396728515625,
-0.0024166107177734375,
-0.005817413330078125,
-0.045440673828125,
-0.0577392578125,
0.0509033203125,
0.00010156631469726562,
0.0499267578125,
0.0179901123046875,
-0.0013074874877929688,
0.0654296875,
-0.019134521484375,
0.081298828125,
0.0322265625,
-0.0673828125,
0.044769287109375,
-0.0343017578125,
0.0297393798828125,
0.027801513671875,
0.0294189453125,
-0.0272369384765625,
-0.0128173828125,
-0.0731201171875,
-0.06427001953125,
0.0692138671875,
0.029205322265625,
-0.0056610107421875,
-0.0009450912475585938,
0.00643157958984375,
-0.016754150390625,
0.0098114013671875,
-0.0673828125,
-0.0411376953125,
-0.04296875,
-0.018310546875,
-0.0038166046142578125,
-0.0022525787353515625,
-0.0034618377685546875,
-0.03411865234375,
0.0565185546875,
0.0086669921875,
0.03094482421875,
0.0202789306640625,
-0.00527191162109375,
0.01329803466796875,
0.01236724853515625,
0.054229736328125,
0.050628662109375,
-0.0171356201171875,
-0.006458282470703125,
0.00624847412109375,
-0.037811279296875,
0.003177642822265625,
0.0017824172973632812,
-0.021636962890625,
-0.002910614013671875,
0.023773193359375,
0.05487060546875,
0.0033779144287109375,
-0.0081329345703125,
0.0426025390625,
-0.00296783447265625,
-0.04052734375,
-0.054901123046875,
0.00762176513671875,
0.017913818359375,
0.0250701904296875,
0.0229644775390625,
0.01421356201171875,
0.005779266357421875,
-0.038848876953125,
0.0042266845703125,
0.032745361328125,
-0.017974853515625,
-0.01080322265625,
0.0650634765625,
0.01425933837890625,
-0.0186767578125,
0.039031982421875,
-0.0185699462890625,
-0.05078125,
0.06048583984375,
0.0394287109375,
0.03521728515625,
-0.026397705078125,
0.01482391357421875,
0.0712890625,
0.0214080810546875,
-0.0028171539306640625,
0.0261383056640625,
0.0032176971435546875,
-0.047882080078125,
-0.0003421306610107422,
-0.03143310546875,
-0.017608642578125,
0.0165863037109375,
-0.03570556640625,
0.0200042724609375,
-0.039581298828125,
-0.0209197998046875,
-0.0089874267578125,
0.03265380859375,
-0.060272216796875,
0.0252227783203125,
-0.0024280548095703125,
0.063720703125,
-0.0618896484375,
0.06842041015625,
0.04376220703125,
-0.054107666015625,
-0.09423828125,
-0.0243072509765625,
0.00894927978515625,
-0.0679931640625,
0.032745361328125,
0.01241302490234375,
0.01261138916015625,
0.017730712890625,
-0.033782958984375,
-0.0855712890625,
0.10858154296875,
-0.005352020263671875,
-0.0321044921875,
0.00653076171875,
-0.005870819091796875,
0.0227203369140625,
0.004302978515625,
0.0452880859375,
0.058258056640625,
0.04632568359375,
0.01169586181640625,
-0.07586669921875,
0.003635406494140625,
-0.041778564453125,
-0.01360321044921875,
0.01384735107421875,
-0.07965087890625,
0.083251953125,
-0.015777587890625,
0.0042877197265625,
0.003139495849609375,
0.04779052734375,
0.0323486328125,
0.026458740234375,
0.034088134765625,
0.0689697265625,
0.0738525390625,
-0.0249786376953125,
0.080322265625,
-0.018768310546875,
0.0421142578125,
0.07049560546875,
0.003383636474609375,
0.04010009765625,
0.0281524658203125,
-0.040557861328125,
0.03485107421875,
0.056884765625,
-0.00328826904296875,
0.0394287109375,
0.0023193359375,
-0.01971435546875,
-0.00331878662109375,
0.01456451416015625,
-0.04815673828125,
0.017822265625,
0.02978515625,
-0.01373291015625,
-0.0023555755615234375,
-0.023162841796875,
0.0172271728515625,
0.002197265625,
-0.0236663818359375,
0.049591064453125,
-0.0104827880859375,
-0.042999267578125,
0.05230712890625,
-0.01678466796875,
0.052490234375,
-0.03924560546875,
0.003055572509765625,
-0.0184783935546875,
0.0224761962890625,
-0.0304107666015625,
-0.072021484375,
0.01036834716796875,
-0.009918212890625,
-0.011962890625,
0.0012216567993164062,
0.030975341796875,
-0.01324462890625,
-0.0305938720703125,
0.0175628662109375,
0.0213775634765625,
0.01514434814453125,
0.015899658203125,
-0.07977294921875,
-0.00908660888671875,
0.016937255859375,
-0.048919677734375,
0.01122283935546875,
0.036102294921875,
-0.01047515869140625,
0.04766845703125,
0.0631103515625,
-0.00856781005859375,
0.0220489501953125,
-0.00696563720703125,
0.06787109375,
-0.055084228515625,
-0.037933349609375,
-0.052734375,
0.0274658203125,
-0.0260009765625,
-0.038238525390625,
0.07586669921875,
0.05084228515625,
0.047576904296875,
-0.007411956787109375,
0.0654296875,
-0.023284912109375,
0.046539306640625,
-0.02679443359375,
0.053497314453125,
-0.053131103515625,
0.004116058349609375,
-0.0204925537109375,
-0.06219482421875,
-0.03424072265625,
0.06890869140625,
-0.03955078125,
0.01320648193359375,
0.0518798828125,
0.064208984375,
0.004360198974609375,
0.007472991943359375,
0.01454925537109375,
0.0108642578125,
0.033111572265625,
0.05712890625,
0.0433349609375,
-0.04791259765625,
0.048004150390625,
-0.034332275390625,
-0.00604248046875,
-0.01435089111328125,
-0.047607421875,
-0.07049560546875,
-0.032135009765625,
-0.0313720703125,
-0.0172882080078125,
-0.00981903076171875,
0.061248779296875,
0.055572509765625,
-0.054840087890625,
-0.0130462646484375,
-0.022125244140625,
0.0143585205078125,
-0.004070281982421875,
-0.017822265625,
0.059234619140625,
-0.01544952392578125,
-0.061370849609375,
0.0141448974609375,
0.00641632080078125,
0.018829345703125,
0.00821685791015625,
-0.030059814453125,
-0.0297088623046875,
-0.00021123886108398438,
0.02392578125,
0.0305328369140625,
-0.06396484375,
-0.0260772705078125,
-0.0072479248046875,
-0.020721435546875,
0.0235443115234375,
0.006134033203125,
-0.0491943359375,
0.0188751220703125,
0.0328369140625,
0.027435302734375,
0.05029296875,
-0.0014333724975585938,
0.004665374755859375,
-0.03155517578125,
0.02398681640625,
-0.001239776611328125,
0.0279388427734375,
0.017822265625,
-0.027099609375,
0.059295654296875,
0.031768798828125,
-0.056488037109375,
-0.0482177734375,
-0.01251220703125,
-0.08709716796875,
-0.0125732421875,
0.076904296875,
-0.0144805908203125,
-0.04058837890625,
-0.014923095703125,
-0.028411865234375,
0.0203704833984375,
-0.04266357421875,
0.036285400390625,
0.048858642578125,
-0.01331329345703125,
-0.023895263671875,
-0.053863525390625,
0.04058837890625,
0.0171661376953125,
-0.05987548828125,
-0.0268707275390625,
0.0276336669921875,
0.027130126953125,
0.0257415771484375,
0.0609130859375,
-0.01390838623046875,
0.0188751220703125,
0.014923095703125,
0.01552581787109375,
-0.0073089599609375,
-0.00101470947265625,
0.0027446746826171875,
-0.003734588623046875,
-0.0097503662109375,
-0.034820556640625
]
] |
timm/vit_large_patch16_224.augreg_in21k | 2023-05-06T00:15:19.000Z | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"dataset:imagenet-21k",
"arxiv:2106.10270",
"arxiv:2010.11929",
"license:apache-2.0",
"region:us"
] | image-classification | timm | null | null | timm/vit_large_patch16_224.augreg_in21k | 0 | 7,386 | timm | 2022-12-22T07:45:11 | ---
tags:
- image-classification
- timm
library_name: timm
license: apache-2.0
datasets:
- imagenet-21k
---
# Model card for vit_large_patch16_224.augreg_in21k
A Vision Transformer (ViT) image classification model. Trained on ImageNet-21k (with additional augmentation and regularization) in JAX by paper authors, ported to PyTorch by Ross Wightman.
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 325.7
- GMACs: 59.7
- Activations (M): 43.8
- Image size: 224 x 224
- **Papers:**
- How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers: https://arxiv.org/abs/2106.10270
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale: https://arxiv.org/abs/2010.11929v2
- **Dataset:** ImageNet-21k
- **Original:** https://github.com/google-research/vision_transformer
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('vit_large_patch16_224.augreg_in21k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'vit_large_patch16_224.augreg_in21k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 197, 1024) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
## Citation
```bibtex
@article{steiner2021augreg,
title={How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers},
author={Steiner, Andreas and Kolesnikov, Alexander and and Zhai, Xiaohua and Wightman, Ross and Uszkoreit, Jakob and Beyer, Lucas},
journal={arXiv preprint arXiv:2106.10270},
year={2021}
}
```
```bibtex
@article{dosovitskiy2020vit,
title={An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale},
author={Dosovitskiy, Alexey and Beyer, Lucas and Kolesnikov, Alexander and Weissenborn, Dirk and Zhai, Xiaohua and Unterthiner, Thomas and Dehghani, Mostafa and Minderer, Matthias and Heigold, Georg and Gelly, Sylvain and Uszkoreit, Jakob and Houlsby, Neil},
journal={ICLR},
year={2021}
}
```
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}
```
| 3,807 | [
[
-0.0396728515625,
-0.030426025390625,
-0.0004210472106933594,
0.007785797119140625,
-0.0262908935546875,
-0.024017333984375,
-0.025604248046875,
-0.037322998046875,
0.01491546630859375,
0.0245819091796875,
-0.037322998046875,
-0.036407470703125,
-0.047515869140625,
0.0016603469848632812,
-0.01332855224609375,
0.07281494140625,
-0.01128387451171875,
0.002288818359375,
-0.015716552734375,
-0.031768798828125,
-0.02349853515625,
-0.02032470703125,
-0.0469970703125,
-0.03143310546875,
0.0301055908203125,
0.00916290283203125,
0.042144775390625,
0.047821044921875,
0.059661865234375,
0.035125732421875,
-0.0086669921875,
0.013641357421875,
-0.0240478515625,
-0.0134735107421875,
0.0193634033203125,
-0.045928955078125,
-0.027984619140625,
0.0166473388671875,
0.0567626953125,
0.0285186767578125,
0.01012420654296875,
0.0260009765625,
0.008636474609375,
0.03643798828125,
-0.0275726318359375,
0.017303466796875,
-0.0396728515625,
0.018585205078125,
-0.0023593902587890625,
-0.00348663330078125,
-0.023529052734375,
-0.0240936279296875,
0.0159149169921875,
-0.03912353515625,
0.044891357421875,
-0.003627777099609375,
0.1036376953125,
0.019775390625,
0.0002930164337158203,
0.0180206298828125,
-0.0335693359375,
0.0576171875,
-0.0452880859375,
0.02984619140625,
0.01255035400390625,
0.0146942138671875,
0.0026226043701171875,
-0.07769775390625,
-0.050079345703125,
-0.01280975341796875,
-0.014373779296875,
0.00759124755859375,
-0.021759033203125,
0.0191497802734375,
0.039154052734375,
0.04718017578125,
-0.038238525390625,
0.0008769035339355469,
-0.04278564453125,
-0.022125244140625,
0.044891357421875,
-0.00251007080078125,
0.01448822021484375,
-0.01117706298828125,
-0.04400634765625,
-0.046051025390625,
-0.0222930908203125,
0.01873779296875,
0.0230255126953125,
0.0031642913818359375,
-0.035186767578125,
0.043670654296875,
0.0033092498779296875,
0.052886962890625,
0.0162811279296875,
-0.0162811279296875,
0.0533447265625,
-0.00833892822265625,
-0.027862548828125,
-0.0218353271484375,
0.08251953125,
0.037506103515625,
0.02972412109375,
-0.0011186599731445312,
-0.01580810546875,
-0.01031494140625,
0.0039215087890625,
-0.08160400390625,
-0.0286102294921875,
0.006938934326171875,
-0.03271484375,
-0.0268402099609375,
0.0252227783203125,
-0.0499267578125,
-0.012298583984375,
-0.01148223876953125,
0.05963134765625,
-0.0310821533203125,
-0.013519287109375,
0.00792694091796875,
-0.01380157470703125,
0.0355224609375,
0.0211181640625,
-0.04278564453125,
0.005985260009765625,
0.0174560546875,
0.07745361328125,
0.002323150634765625,
-0.03662109375,
-0.022247314453125,
-0.0305023193359375,
-0.02447509765625,
0.038360595703125,
-0.0047607421875,
-0.0084686279296875,
-0.0098724365234375,
0.0311431884765625,
-0.0190887451171875,
-0.042236328125,
0.0258026123046875,
-0.017486572265625,
0.0257415771484375,
0.0098724365234375,
-0.015045166015625,
-0.031585693359375,
0.0215911865234375,
-0.03253173828125,
0.089599609375,
0.024871826171875,
-0.06646728515625,
0.0294342041015625,
-0.03399658203125,
-0.006893157958984375,
-0.01055908203125,
0.0016155242919921875,
-0.0843505859375,
0.004375457763671875,
0.02490234375,
0.045745849609375,
-0.0167388916015625,
0.00147247314453125,
-0.0322265625,
-0.025604248046875,
0.02947998046875,
-0.0213470458984375,
0.06707763671875,
0.0007772445678710938,
-0.02435302734375,
0.0216827392578125,
-0.04425048828125,
0.0050201416015625,
0.0306396484375,
-0.020416259765625,
0.0018053054809570312,
-0.0479736328125,
0.014251708984375,
0.01641845703125,
0.018829345703125,
-0.050018310546875,
0.03021240234375,
-0.023651123046875,
0.033843994140625,
0.050384521484375,
-0.007587432861328125,
0.0264434814453125,
-0.0242462158203125,
0.0236053466796875,
0.0167388916015625,
0.02978515625,
-0.0116424560546875,
-0.046844482421875,
-0.0775146484375,
-0.036865234375,
0.0263671875,
0.03131103515625,
-0.0498046875,
0.044342041015625,
-0.027801513671875,
-0.054962158203125,
-0.042144775390625,
0.002552032470703125,
0.033172607421875,
0.036041259765625,
0.039093017578125,
-0.040985107421875,
-0.041351318359375,
-0.07208251953125,
-0.00884246826171875,
-0.0026760101318359375,
-0.0007982254028320312,
0.01727294921875,
0.04931640625,
-0.0225372314453125,
0.060821533203125,
-0.0364990234375,
-0.0262298583984375,
-0.01517486572265625,
0.003444671630859375,
0.02703857421875,
0.055450439453125,
0.048980712890625,
-0.04840087890625,
-0.03143310546875,
-0.01012420654296875,
-0.062164306640625,
0.01055908203125,
-0.0034465789794921875,
-0.0124053955078125,
0.0118408203125,
0.0184326171875,
-0.05462646484375,
0.056610107421875,
0.0169830322265625,
-0.0269317626953125,
0.032012939453125,
-0.01953125,
0.005916595458984375,
-0.09051513671875,
0.0020847320556640625,
0.029327392578125,
-0.0199432373046875,
-0.032989501953125,
0.0015687942504882812,
0.00943756103515625,
-0.004222869873046875,
-0.02703857421875,
0.042266845703125,
-0.038909912109375,
-0.0061187744140625,
-0.004100799560546875,
-0.024810791015625,
0.00691986083984375,
0.052703857421875,
-0.0037994384765625,
0.0406494140625,
0.052459716796875,
-0.034698486328125,
0.049468994140625,
0.038604736328125,
-0.0169830322265625,
0.03662109375,
-0.05487060546875,
0.009063720703125,
-0.0017652511596679688,
0.0189361572265625,
-0.07415771484375,
-0.019989013671875,
0.025299072265625,
-0.054931640625,
0.050079345703125,
-0.038787841796875,
-0.033966064453125,
-0.049468994140625,
-0.0304412841796875,
0.032012939453125,
0.056060791015625,
-0.059661865234375,
0.045440673828125,
0.00731658935546875,
0.0250396728515625,
-0.0435791015625,
-0.0718994140625,
-0.01275634765625,
-0.0290679931640625,
-0.051910400390625,
0.033538818359375,
0.00482177734375,
0.0120391845703125,
0.00311279296875,
-0.0023899078369140625,
0.00006896257400512695,
-0.0179595947265625,
0.033172607421875,
0.0294647216796875,
-0.015869140625,
-0.0022945404052734375,
-0.0287933349609375,
-0.01512908935546875,
0.0006537437438964844,
-0.026885986328125,
0.041717529296875,
-0.02252197265625,
-0.016937255859375,
-0.05682373046875,
-0.0205841064453125,
0.037994384765625,
-0.0253448486328125,
0.055999755859375,
0.08990478515625,
-0.033966064453125,
0.005382537841796875,
-0.04693603515625,
-0.028778076171875,
-0.0367431640625,
0.037445068359375,
-0.0210418701171875,
-0.0341796875,
0.056121826171875,
0.0122528076171875,
0.0078277587890625,
0.0584716796875,
0.0328369140625,
0.005645751953125,
0.0643310546875,
0.049713134765625,
0.0123138427734375,
0.0662841796875,
-0.0732421875,
-0.007266998291015625,
-0.0718994140625,
-0.0287933349609375,
-0.0194549560546875,
-0.0401611328125,
-0.053192138671875,
-0.04010009765625,
0.031951904296875,
0.004856109619140625,
-0.0242462158203125,
0.037841796875,
-0.06585693359375,
0.0128021240234375,
0.0546875,
0.037872314453125,
-0.006702423095703125,
0.03253173828125,
-0.0108642578125,
-0.003856658935546875,
-0.054656982421875,
-0.003551483154296875,
0.08233642578125,
0.0355224609375,
0.06072998046875,
-0.0207977294921875,
0.04962158203125,
-0.0201873779296875,
0.0202178955078125,
-0.06011962890625,
0.041168212890625,
0.0004782676696777344,
-0.0304412841796875,
-0.01018524169921875,
-0.03192138671875,
-0.08074951171875,
0.0157928466796875,
-0.027435302734375,
-0.060882568359375,
0.0262298583984375,
0.01415252685546875,
-0.01377105712890625,
0.05157470703125,
-0.0645751953125,
0.070556640625,
-0.003444671630859375,
-0.037872314453125,
0.006122589111328125,
-0.050628662109375,
0.0162506103515625,
0.01486968994140625,
-0.0251922607421875,
0.012237548828125,
0.0184783935546875,
0.07525634765625,
-0.04400634765625,
0.063232421875,
-0.028472900390625,
0.0247802734375,
0.03387451171875,
-0.0182952880859375,
0.030731201171875,
0.0031375885009765625,
0.01317596435546875,
0.0226593017578125,
-0.0022258758544921875,
-0.0287628173828125,
-0.037628173828125,
0.039306640625,
-0.0792236328125,
-0.030609130859375,
-0.0406494140625,
-0.046600341796875,
0.00818634033203125,
0.006664276123046875,
0.047943115234375,
0.047882080078125,
0.0234375,
0.03265380859375,
0.050628662109375,
-0.02789306640625,
0.0310821533203125,
-0.00035881996154785156,
-0.01496124267578125,
-0.043670654296875,
0.07183837890625,
0.0165252685546875,
0.0122833251953125,
0.01209259033203125,
0.01763916015625,
-0.025787353515625,
-0.0364990234375,
-0.02508544921875,
0.032989501953125,
-0.0509033203125,
-0.037322998046875,
-0.0443115234375,
-0.041259765625,
-0.0251617431640625,
0.0014047622680664062,
-0.030029296875,
-0.024658203125,
-0.02532958984375,
0.00939178466796875,
0.06524658203125,
0.038909912109375,
-0.011077880859375,
0.037689208984375,
-0.04266357421875,
0.015228271484375,
0.023040771484375,
0.0384521484375,
-0.01306915283203125,
-0.075439453125,
-0.0269775390625,
0.002040863037109375,
-0.03936767578125,
-0.053436279296875,
0.035491943359375,
0.01502227783203125,
0.0313720703125,
0.0284881591796875,
-0.018157958984375,
0.06658935546875,
-0.005077362060546875,
0.04473876953125,
0.0265045166015625,
-0.03997802734375,
0.0347900390625,
-0.01018524169921875,
0.0109710693359375,
0.011077880859375,
0.01024627685546875,
-0.0236663818359375,
-0.006320953369140625,
-0.081298828125,
-0.059661865234375,
0.05596923828125,
0.016082763671875,
0.006343841552734375,
0.034637451171875,
0.04351806640625,
-0.00829315185546875,
0.005374908447265625,
-0.06768798828125,
-0.0190582275390625,
-0.02630615234375,
-0.02410888671875,
-0.011260986328125,
-0.00197601318359375,
-0.0029964447021484375,
-0.058746337890625,
0.047760009765625,
-0.00605010986328125,
0.06121826171875,
0.033843994140625,
-0.0149078369140625,
-0.0109710693359375,
-0.0302581787109375,
0.0262908935546875,
0.0208740234375,
-0.019866943359375,
0.00443267822265625,
0.0201568603515625,
-0.0556640625,
-0.00499725341796875,
0.0257568359375,
-0.00470733642578125,
0.002582550048828125,
0.038482666015625,
0.08099365234375,
-0.010009765625,
-0.0029850006103515625,
0.043609619140625,
-0.00555419921875,
-0.03131103515625,
-0.0239410400390625,
0.0050811767578125,
-0.0172271728515625,
0.0298309326171875,
0.0240936279296875,
0.0283660888671875,
-0.0107879638671875,
-0.01090240478515625,
0.00937652587890625,
0.0406494140625,
-0.0399169921875,
-0.0267791748046875,
0.048187255859375,
-0.01409912109375,
-0.00600433349609375,
0.0604248046875,
-0.0021915435791015625,
-0.0423583984375,
0.06646728515625,
0.0250701904296875,
0.0732421875,
-0.009124755859375,
-0.0025272369384765625,
0.059478759765625,
0.026519775390625,
-0.00499725341796875,
0.009979248046875,
0.01074981689453125,
-0.056304931640625,
-0.01183319091796875,
-0.04888916015625,
0.003177642822265625,
0.02655029296875,
-0.040557861328125,
0.0303497314453125,
-0.040374755859375,
-0.0279693603515625,
0.004222869873046875,
0.0182037353515625,
-0.0750732421875,
0.0216217041015625,
-0.0003941059112548828,
0.055877685546875,
-0.061920166015625,
0.044708251953125,
0.0640869140625,
-0.0506591796875,
-0.07281494140625,
-0.01183319091796875,
-0.0150909423828125,
-0.06964111328125,
0.033203125,
0.033966064453125,
0.0157470703125,
0.01776123046875,
-0.060699462890625,
-0.047454833984375,
0.09576416015625,
0.02685546875,
-0.01302337646484375,
0.0100555419921875,
-0.00494384765625,
0.0293426513671875,
-0.01824951171875,
0.035430908203125,
0.01136016845703125,
0.0311737060546875,
0.0188446044921875,
-0.05352783203125,
0.0064239501953125,
-0.02337646484375,
0.0127410888671875,
0.0172119140625,
-0.062164306640625,
0.07318115234375,
-0.031890869140625,
-0.003856658935546875,
0.01482391357421875,
0.047515869140625,
0.0078277587890625,
0.00460052490234375,
0.040191650390625,
0.06622314453125,
0.0305023193359375,
-0.032257080078125,
0.0687255859375,
-0.00995635986328125,
0.054046630859375,
0.035675048828125,
0.036865234375,
0.0345458984375,
0.035491943359375,
-0.024322509765625,
0.024200439453125,
0.07537841796875,
-0.045501708984375,
0.0175628662109375,
0.01025390625,
0.0037899017333984375,
-0.019683837890625,
0.003467559814453125,
-0.03790283203125,
0.039093017578125,
0.0151824951171875,
-0.044647216796875,
-0.005741119384765625,
0.01361083984375,
-0.01071929931640625,
-0.0276641845703125,
-0.015716552734375,
0.04541015625,
0.0001474618911743164,
-0.03350830078125,
0.064697265625,
-0.0013713836669921875,
0.058868408203125,
-0.0304412841796875,
-0.002017974853515625,
-0.016845703125,
0.034423828125,
-0.0297393798828125,
-0.060882568359375,
0.0104522705078125,
-0.0185699462890625,
-0.004505157470703125,
0.0050811767578125,
0.04931640625,
-0.03009033203125,
-0.043975830078125,
0.00695037841796875,
0.0213165283203125,
0.0229339599609375,
-0.006305694580078125,
-0.0750732421875,
-0.00008767843246459961,
0.000682830810546875,
-0.04449462890625,
0.015655517578125,
0.0321044921875,
-0.00016057491302490234,
0.05120849609375,
0.052520751953125,
-0.004505157470703125,
0.0149688720703125,
-0.00897979736328125,
0.07037353515625,
-0.032958984375,
-0.028411865234375,
-0.0582275390625,
0.046234130859375,
-0.00884246826171875,
-0.045318603515625,
0.050689697265625,
0.046844482421875,
0.06658935546875,
-0.01020050048828125,
0.0345458984375,
-0.00974273681640625,
0.0023040771484375,
-0.026885986328125,
0.04180908203125,
-0.0533447265625,
-0.00811004638671875,
-0.0233306884765625,
-0.07177734375,
-0.03143310546875,
0.07086181640625,
-0.029296875,
0.03448486328125,
0.03961181640625,
0.072998046875,
-0.0250091552734375,
-0.030120849609375,
0.01222991943359375,
0.0178375244140625,
0.01212310791015625,
0.030853271484375,
0.045928955078125,
-0.06365966796875,
0.037872314453125,
-0.0413818359375,
-0.0151824951171875,
-0.0214385986328125,
-0.036041259765625,
-0.07965087890625,
-0.06243896484375,
-0.04193115234375,
-0.0540771484375,
-0.0153656005859375,
0.06298828125,
0.07000732421875,
-0.040740966796875,
-0.004978179931640625,
-0.01142120361328125,
0.0014190673828125,
-0.0217132568359375,
-0.0177459716796875,
0.041015625,
-0.01007080078125,
-0.0594482421875,
-0.028167724609375,
-0.0009317398071289062,
0.037078857421875,
-0.01522064208984375,
-0.00927734375,
-0.00894927978515625,
-0.0253143310546875,
0.01904296875,
0.023468017578125,
-0.0513916015625,
-0.018646240234375,
-0.00740814208984375,
-0.0023345947265625,
0.036041259765625,
0.0287933349609375,
-0.053253173828125,
0.04095458984375,
0.041778564453125,
0.0265045166015625,
0.0650634765625,
-0.0156097412109375,
0.0099945068359375,
-0.060882568359375,
0.045074462890625,
-0.0036258697509765625,
0.040435791015625,
0.04010009765625,
-0.019378662109375,
0.04205322265625,
0.042999267578125,
-0.03717041015625,
-0.0645751953125,
-0.000804901123046875,
-0.08148193359375,
0.00928497314453125,
0.07568359375,
-0.018218994140625,
-0.0380859375,
0.028411865234375,
-0.0175018310546875,
0.053070068359375,
-0.0061798095703125,
0.0374755859375,
0.01495361328125,
0.0087738037109375,
-0.045166015625,
-0.036407470703125,
0.03643798828125,
0.00839996337890625,
-0.04022216796875,
-0.026947021484375,
0.00643157958984375,
0.039581298828125,
0.028411865234375,
0.02001953125,
-0.011749267578125,
0.01276397705078125,
0.0024967193603515625,
0.0416259765625,
-0.02685546875,
-0.010345458984375,
-0.0302886962890625,
-0.012237548828125,
-0.0038299560546875,
-0.045928955078125
]
] |
TheBloke/llama-30b-supercot-SuperHOT-8K-fp16 | 2023-07-09T20:24:56.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"custom_code",
"license:other",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/llama-30b-supercot-SuperHOT-8K-fp16 | 4 | 7,381 | transformers | 2023-06-29T00:15:45 | ---
inference: false
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/theblokeai">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 -->
# Ausboss' Llama 30B SuperCOT fp16
This is fp16 pytorch format model files for [Ausboss' Llama 30B SuperCOT](https://huggingface.co/ausboss/llama-30b-supercot) merged with [Kaio Ken's SuperHOT 8K](https://huggingface.co/kaiokendev/superhot-30b-8k-no-rlhf-test).
[Kaio Ken's SuperHOT 30b LoRA](https://huggingface.co/kaiokendev/superhot-30b-8k-no-rlhf-test) is merged on to the base model, and then 8K context can be achieved during inference by using `trust_remote_code=True`.
Note that `config.json` has been set to a sequence length of 8192. This can be modified to 4096 if you want to try with a smaller sequence length.
## Repositories available
* [4-bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/llama-30b-supercot-SuperHOT-8K-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGML models for CPU inference](https://huggingface.co/TheBloke/llama-30b-supercot-SuperHOT-8K-GGML)
* [Unquantised SuperHOT fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/TheBloke/llama-30b-supercot-SuperHOT-8K-fp16)
* [Unquantised base fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/ausboss/llama-30b-supercot)
## How to use this model from Python code
First make sure you have Einops installed:
```
pip3 install auto-gptq
```
Then run the following code. `config.json` has been default to a sequence length of 8192, but you can also configure this in your Python code.
The provided modelling code, activated with `trust_remote_code=True` will automatically set the `scale` parameter from the configured `max_position_embeddings`. Eg for 8192, `scale` is set to `4`.
```python
from transformers import AutoConfig, AutoTokenizer, AutoModelForCausalLM, pipeline
import argparse
model_name_or_path = "TheBloke/llama-30b-supercot-SuperHOT-8K-fp16"
use_triton = False
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
config = AutoConfig.from_pretrained(model_name_or_path, trust_remote_code=True)
# Change this to the sequence length you want
config.max_position_embeddings = 8192
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
config=config,
trust_remote_code=True,
device_map='auto')
# Note: check to confirm if this is correct prompt template is correct for this model!
prompt = "Tell me about AI"
prompt_template=f'''USER: {prompt}
ASSISTANT:'''
print("\n\n*** Generate:")
input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, max_new_tokens=512)
print(tokenizer.decode(output[0]))
# Inference can also be done using transformers' pipeline
print("*** Pipeline:")
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
max_new_tokens=512,
temperature=0.7,
top_p=0.95,
repetition_penalty=1.15
)
print(pipe(prompt_template)[0]['generated_text'])
```
## Using other UIs: monkey patch
Provided in the repo is `llama_rope_scaled_monkey_patch.py`, written by @kaiokendev.
It can be theoretically be added to any Python UI or custom code to enable the same result as `trust_remote_code=True`. I have not tested this, and it should be superseded by using `trust_remote_code=True`, but I include it for completeness and for interest.
<!-- 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/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**: Luke from CarbonQuill, Aemon Algiz, Dmitriy Samsonov.
**Patreon special mentions**: zynix , ya boyyy, Trenton Dambrowitz, Imad Khwaja, Alps Aficionado, chris gileta, John Detwiler, Willem Michiel, RoA, Mano Prime, Rainer Wilmers, Fred von Graf, Matthew Berman, Ghost , Nathan LeClaire, Iucharbius , Ai Maven, Illia Dulskyi, Joseph William Delisle, Space Cruiser, Lone Striker, Karl Bernard, Eugene Pentland, Greatston Gnanesh, Jonathan Leane, Randy H, Pierre Kircher, Willian Hasse, Stephen Murray, Alex , terasurfer , Edmond Seymore, Oscar Rangel, Luke Pendergrass, Asp the Wyvern, Junyu Yang, David Flickinger, Luke, Spiking Neurons AB, subjectnull, Pyrater, Nikolai Manek, senxiiz, Ajan Kanaga, Johann-Peter Hartmann, Artur Olbinski, Kevin Schuppel, Derek Yates, Kalila, K, Talal Aujan, Khalefa Al-Ahmad, Gabriel Puliatti, John Villwock, WelcomeToTheClub, Daniel P. Andersen, Preetika Verma, Deep Realms, Fen Risland, trip7s trip, webtim, Sean Connelly, Michael Levine, Chris McCloskey, biorpg, vamX, Viktor Bowallius, Cory Kujawski.
Thank you to all my generous patrons and donaters!
<!-- footer end -->
# Original model card: Kaio Ken's SuperHOT 8K
### SuperHOT Prototype 2 w/ 8K Context
This is a second prototype of SuperHOT, this time 30B with 8K context and no RLHF, using the same technique described in [the github blog](https://kaiokendev.github.io/til#extending-context-to-8k).
Tests have shown that the model does indeed leverage the extended context at 8K.
You will need to **use either the monkeypatch** or, if you are already using the monkeypatch, **change the scaling factor to 0.25 and the maximum sequence length to 8192**
#### Looking for Merged & Quantized Models?
- 30B 4-bit CUDA: [tmpupload/superhot-30b-8k-4bit-safetensors](https://huggingface.co/tmpupload/superhot-30b-8k-4bit-safetensors)
- 30B 4-bit CUDA 128g: [tmpupload/superhot-30b-8k-4bit-128g-safetensors](https://huggingface.co/tmpupload/superhot-30b-8k-4bit-128g-safetensors)
#### Training Details
I trained the LoRA with the following configuration:
- 1200 samples (~400 samples over 2048 sequence length)
- learning rate of 3e-4
- 3 epochs
- The exported modules are:
- q_proj
- k_proj
- v_proj
- o_proj
- no bias
- Rank = 4
- Alpha = 8
- no dropout
- weight decay of 0.1
- AdamW beta1 of 0.9 and beta2 0.99, epsilon of 1e-5
- Trained on 4-bit base model
# Original model card: Ausboss' Llama 30B SuperCOT
Merge of [huggyllama/llama-30b](https://huggingface.co/huggyllama/llama-30b) + [kaiokendev/SuperCOT-LoRA](https://huggingface.co/kaiokendev/SuperCOT-LoRA/edit/main/README.md)
Supercot was trained to work with langchain prompting.
Load up locally in my custom LLM notebook that uses the Oobabooga modules to load up models: https://github.com/ausboss/Local-LLM-Langchain
Then you can add cells from of these other notebooks for testing: https://github.com/gkamradt/langchain-tutorials
# From Koikendev Lora page
### Compatibility
This LoRA is compatible with any 7B, 13B or 30B 4-bit quantized LLaMa model, including ggml quantized converted bins
### Prompting
You should prompt the LoRA the same way you would prompt Alpaca or Alpacino:
```
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:
<instruction>
### Input:
<any additional context. Remove this if it's not neccesary>
### Response:
<make sure to leave a single new-line here for optimal results>
```
Remember that with lower parameter sizes, the structure of the prompt becomes more important. The same prompt worded differently can give wildly different answers. Consider using the following suggestion suffixes to improve output quality:
- "Think through this step by step"
- "Let's think about this logically"
- "Explain your reasoning"
- "Provide details to support your answer"
- "Compare and contrast your answer with alternatives"
### Coming Soon
- Tweet fix for 13B and 7B - lower model sizes seem to be extremely sensitive to hashtags at the end of training data responses, especially at longer cutoffs
| 9,115 | [
[
-0.03094482421875,
-0.057342529296875,
0.014190673828125,
-0.00736236572265625,
-0.0258331298828125,
-0.01088714599609375,
0.0023746490478515625,
-0.046356201171875,
0.0293426513671875,
0.0126953125,
-0.0469970703125,
-0.03179931640625,
-0.035614013671875,
0.00140380859375,
-0.0234222412109375,
0.07904052734375,
0.01256561279296875,
-0.007549285888671875,
0.004215240478515625,
-0.0034313201904296875,
-0.022613525390625,
-0.03509521484375,
-0.046234130859375,
-0.023345947265625,
0.024017333984375,
0.0118408203125,
0.061737060546875,
0.042266845703125,
0.032958984375,
0.0244903564453125,
-0.0262451171875,
0.0163421630859375,
-0.031402587890625,
-0.0156707763671875,
0.0017452239990234375,
-0.03411865234375,
-0.052490234375,
-0.0093994140625,
0.03900146484375,
0.006725311279296875,
-0.01092529296875,
0.0214080810546875,
-0.014495849609375,
0.0203704833984375,
-0.0278167724609375,
0.01355743408203125,
-0.0350341796875,
0.005832672119140625,
-0.0011615753173828125,
0.01617431640625,
-0.0176544189453125,
-0.01242828369140625,
-0.004482269287109375,
-0.07086181640625,
0.0167388916015625,
0.00675201416015625,
0.09796142578125,
0.0074462890625,
-0.023651123046875,
0.014801025390625,
-0.031494140625,
0.0501708984375,
-0.074462890625,
0.0199737548828125,
0.014068603515625,
0.01953125,
-0.0115509033203125,
-0.0655517578125,
-0.054534912109375,
-0.0208587646484375,
-0.00542449951171875,
0.019561767578125,
-0.024627685546875,
-0.0077362060546875,
0.034423828125,
0.032501220703125,
-0.036712646484375,
-0.0010471343994140625,
-0.0255126953125,
-0.0171051025390625,
0.053497314453125,
0.01540374755859375,
0.028564453125,
-0.0168914794921875,
-0.018951416015625,
-0.0263214111328125,
-0.04620361328125,
0.0061492919921875,
0.0194549560546875,
0.016571044921875,
-0.047607421875,
0.037017822265625,
-0.0036678314208984375,
0.04034423828125,
0.0199737548828125,
-0.0019741058349609375,
0.0158538818359375,
-0.039794921875,
-0.03228759765625,
-0.014495849609375,
0.08172607421875,
0.030120849609375,
-0.0015153884887695312,
0.026580810546875,
0.0013580322265625,
-0.0048370361328125,
-0.009307861328125,
-0.06683349609375,
-0.0256805419921875,
0.03326416015625,
-0.02813720703125,
-0.0199127197265625,
-0.00019097328186035156,
-0.038482666015625,
-0.01476287841796875,
0.005603790283203125,
0.0330810546875,
-0.0435791015625,
-0.0328369140625,
0.005847930908203125,
-0.0240631103515625,
0.034698486328125,
0.0269622802734375,
-0.07586669921875,
0.0081634521484375,
0.03326416015625,
0.0556640625,
0.03656005859375,
-0.0246734619140625,
-0.0201416015625,
0.007266998291015625,
-0.0273895263671875,
0.040496826171875,
-0.0159912109375,
-0.03594970703125,
-0.0245513916015625,
0.018951416015625,
0.006381988525390625,
-0.0267791748046875,
0.01092529296875,
-0.031890869140625,
0.01123809814453125,
-0.0194854736328125,
-0.034332275390625,
-0.01094818115234375,
0.0171356201171875,
-0.035888671875,
0.071533203125,
0.0273895263671875,
-0.05657958984375,
0.01294708251953125,
-0.05047607421875,
-0.01500701904296875,
0.0111236572265625,
-0.0130615234375,
-0.042205810546875,
-0.00330352783203125,
0.022186279296875,
0.027069091796875,
-0.0312347412109375,
0.01617431640625,
-0.02764892578125,
-0.025390625,
0.021881103515625,
-0.0254669189453125,
0.0831298828125,
0.01302337646484375,
-0.04119873046875,
0.006114959716796875,
-0.049346923828125,
-0.004913330078125,
0.03289794921875,
-0.017181396484375,
0.0081787109375,
-0.036834716796875,
0.0106048583984375,
0.0002658367156982422,
0.0234832763671875,
-0.04376220703125,
0.0278778076171875,
-0.025054931640625,
0.0556640625,
0.05889892578125,
-0.0038814544677734375,
0.0157928466796875,
-0.0234832763671875,
0.0386962890625,
0.004474639892578125,
0.040435791015625,
-0.01120758056640625,
-0.046783447265625,
-0.059051513671875,
-0.024017333984375,
0.0357666015625,
0.03350830078125,
-0.04815673828125,
0.04547119140625,
-0.0158843994140625,
-0.04974365234375,
-0.03314208984375,
-0.01062774658203125,
0.03485107421875,
0.0347900390625,
0.040679931640625,
-0.03192138671875,
-0.04315185546875,
-0.06085205078125,
0.01251220703125,
-0.01515960693359375,
-0.0091705322265625,
0.040252685546875,
0.053863525390625,
-0.0228424072265625,
0.054229736328125,
-0.03704833984375,
-0.0175933837890625,
-0.0276031494140625,
-0.0177764892578125,
0.036468505859375,
0.05731201171875,
0.04193115234375,
-0.050506591796875,
-0.032196044921875,
0.01280975341796875,
-0.057220458984375,
0.01548004150390625,
-0.0034618377685546875,
-0.0171356201171875,
0.01468658447265625,
0.023651123046875,
-0.0811767578125,
0.04022216796875,
0.02764892578125,
-0.027587890625,
0.057647705078125,
-0.0233917236328125,
0.00537109375,
-0.09765625,
0.0156707763671875,
0.0035381317138671875,
-0.01849365234375,
-0.036407470703125,
0.022430419921875,
0.0026760101318359375,
-0.00424957275390625,
-0.052642822265625,
0.049224853515625,
-0.0312042236328125,
0.0007410049438476562,
0.007007598876953125,
-0.00856781005859375,
0.016571044921875,
0.0291748046875,
-0.002864837646484375,
0.042724609375,
0.050018310546875,
-0.0279998779296875,
0.040740966796875,
0.0267333984375,
-0.0177001953125,
0.0258941650390625,
-0.07110595703125,
0.006671905517578125,
0.0167083740234375,
0.05255126953125,
-0.06536865234375,
-0.0300445556640625,
0.031982421875,
-0.05133056640625,
0.031402587890625,
-0.0202484130859375,
-0.0254058837890625,
-0.04132080078125,
-0.0211944580078125,
0.036102294921875,
0.05291748046875,
-0.035308837890625,
0.049713134765625,
0.02178955078125,
0.016845703125,
-0.05169677734375,
-0.0718994140625,
-0.004245758056640625,
-0.0190582275390625,
-0.038177490234375,
0.03289794921875,
-0.00572967529296875,
0.00038623809814453125,
-0.003467559814453125,
-0.0027313232421875,
0.0074462890625,
-0.007717132568359375,
0.0288848876953125,
0.049285888671875,
-0.0248260498046875,
-0.022857666015625,
-0.00803375244140625,
-0.00653076171875,
-0.01067352294921875,
-0.027130126953125,
0.05224609375,
-0.027435302734375,
-0.00611114501953125,
-0.0699462890625,
0.01247406005859375,
0.04754638671875,
-0.006866455078125,
0.06085205078125,
0.049041748046875,
-0.0287322998046875,
0.006282806396484375,
-0.03302001953125,
-0.0242919921875,
-0.0396728515625,
0.0236053466796875,
-0.02545166015625,
-0.059356689453125,
0.05523681640625,
0.026092529296875,
0.0092926025390625,
0.049407958984375,
0.0250244140625,
-0.00655364990234375,
0.0771484375,
0.0418701171875,
-0.01316070556640625,
0.044189453125,
-0.0794677734375,
0.0103912353515625,
-0.06854248046875,
-0.0152587890625,
-0.01520538330078125,
-0.0257415771484375,
-0.046356201171875,
-0.0391845703125,
0.0245208740234375,
0.01081085205078125,
-0.050811767578125,
0.046783447265625,
-0.038055419921875,
0.03570556640625,
0.042724609375,
0.0240325927734375,
0.0022754669189453125,
-0.0020542144775390625,
-0.0158233642578125,
0.01486968994140625,
-0.069580078125,
-0.0152435302734375,
0.07867431640625,
0.0272064208984375,
0.046539306640625,
0.0011873245239257812,
0.050872802734375,
-0.0020847320556640625,
0.0223846435546875,
-0.041046142578125,
0.049285888671875,
-0.01129150390625,
-0.062347412109375,
-0.020965576171875,
-0.02783203125,
-0.058837890625,
0.006229400634765625,
-0.01346588134765625,
-0.055816650390625,
0.0284423828125,
0.0160369873046875,
-0.06329345703125,
0.034088134765625,
-0.040863037109375,
0.06695556640625,
-0.00482177734375,
-0.029144287109375,
-0.005046844482421875,
-0.0645751953125,
0.0298004150390625,
0.021240234375,
0.00966644287109375,
-0.0092926025390625,
0.006900787353515625,
0.0538330078125,
-0.0604248046875,
0.0689697265625,
-0.0089263916015625,
0.0008840560913085938,
0.044403076171875,
0.00424957275390625,
0.0305938720703125,
0.02044677734375,
-0.0013723373413085938,
0.034393310546875,
0.0088043212890625,
-0.0304718017578125,
-0.0281982421875,
0.051025390625,
-0.08050537109375,
-0.046630859375,
-0.0225830078125,
-0.0482177734375,
0.0195770263671875,
0.019989013671875,
0.037109375,
0.04364013671875,
0.00751495361328125,
0.02545166015625,
0.0309600830078125,
-0.0226593017578125,
0.03900146484375,
0.02362060546875,
-0.01641845703125,
-0.042877197265625,
0.0654296875,
0.01105499267578125,
0.025177001953125,
0.024078369140625,
0.0211029052734375,
-0.0289459228515625,
-0.020263671875,
-0.0496826171875,
0.028900146484375,
-0.04547119140625,
-0.040557861328125,
-0.03759765625,
-0.0303192138671875,
-0.052703857421875,
-0.0180511474609375,
-0.041717529296875,
-0.03448486328125,
-0.04315185546875,
-0.01139068603515625,
0.0528564453125,
0.031585693359375,
-0.0188140869140625,
0.02764892578125,
-0.0545654296875,
0.018280029296875,
0.0276641845703125,
-0.0015716552734375,
0.00473785400390625,
-0.06878662109375,
-0.01445770263671875,
0.01064300537109375,
-0.033721923828125,
-0.040069580078125,
0.047119140625,
0.00988006591796875,
0.0274505615234375,
0.018890380859375,
-0.0006456375122070312,
0.07159423828125,
-0.0196380615234375,
0.08258056640625,
0.0294189453125,
-0.07220458984375,
0.05072021484375,
-0.034210205078125,
0.0308074951171875,
0.0226287841796875,
0.03533935546875,
-0.0203399658203125,
-0.03143310546875,
-0.07635498046875,
-0.0709228515625,
0.046356201171875,
0.006175994873046875,
0.017913818359375,
-0.004299163818359375,
0.041290283203125,
-0.01068115234375,
-0.003604888916015625,
-0.07220458984375,
-0.025238037109375,
-0.0219268798828125,
-0.00301361083984375,
-0.00217437744140625,
0.0003108978271484375,
-0.0190887451171875,
-0.039337158203125,
0.0631103515625,
-0.01555633544921875,
0.0369873046875,
0.03497314453125,
0.00434112548828125,
-0.0102996826171875,
-0.0022220611572265625,
0.033477783203125,
0.048919677734375,
-0.0194854736328125,
-0.005855560302734375,
0.019683837890625,
-0.03607177734375,
0.00865936279296875,
0.01255035400390625,
-0.0180206298828125,
-0.0125885009765625,
0.0281829833984375,
0.064208984375,
0.009002685546875,
-0.0323486328125,
0.0298614501953125,
-0.009368896484375,
-0.0210418701171875,
-0.00646209716796875,
0.022247314453125,
0.01971435546875,
0.034759521484375,
0.0226898193359375,
-0.0059814453125,
-0.00844573974609375,
-0.0477294921875,
0.00730133056640625,
0.035858154296875,
-0.018798828125,
-0.025146484375,
0.078857421875,
0.000027835369110107422,
-0.02392578125,
0.058349609375,
0.0020122528076171875,
-0.034332275390625,
0.08416748046875,
0.055633544921875,
0.05474853515625,
-0.00852203369140625,
0.0079498291015625,
0.036285400390625,
0.01177215576171875,
-0.00446319580078125,
0.01959228515625,
-0.0008034706115722656,
-0.049530029296875,
-0.00646209716796875,
-0.045318603515625,
-0.032196044921875,
0.0174407958984375,
-0.045013427734375,
0.0282745361328125,
-0.061614990234375,
-0.028289794921875,
0.0030612945556640625,
0.006687164306640625,
-0.046905517578125,
0.01904296875,
0.01273345947265625,
0.08221435546875,
-0.0546875,
0.06768798828125,
0.0386962890625,
-0.047515869140625,
-0.08343505859375,
-0.0171356201171875,
-0.0025806427001953125,
-0.06439208984375,
0.0312347412109375,
0.01873779296875,
0.0034618377685546875,
0.020721435546875,
-0.059326171875,
-0.058746337890625,
0.11334228515625,
0.0295257568359375,
-0.042816162109375,
0.004390716552734375,
-0.00676727294921875,
0.03741455078125,
-0.03369140625,
0.051177978515625,
0.04132080078125,
0.0118255615234375,
0.009674072265625,
-0.07110595703125,
0.01739501953125,
-0.02764892578125,
0.009674072265625,
-0.0002987384796142578,
-0.090576171875,
0.069091796875,
-0.032012939453125,
-0.00151824951171875,
0.023101806640625,
0.0540771484375,
0.039215087890625,
0.001232147216796875,
0.0243072509765625,
0.0550537109375,
0.040863037109375,
0.0007238388061523438,
0.0865478515625,
-0.035797119140625,
0.04193115234375,
0.07061767578125,
-0.003509521484375,
0.06634521484375,
0.0235595703125,
-0.034149169921875,
0.04327392578125,
0.060333251953125,
-0.006481170654296875,
0.02386474609375,
0.003757476806640625,
-0.01654052734375,
-0.0166015625,
-0.018463134765625,
-0.06695556640625,
0.0168304443359375,
0.0290374755859375,
-0.0244598388671875,
0.01363372802734375,
-0.01088714599609375,
-0.005428314208984375,
-0.04058837890625,
-0.01459503173828125,
0.0322265625,
0.0198974609375,
-0.026763916015625,
0.06878662109375,
-0.00769805908203125,
0.05731201171875,
-0.05499267578125,
0.003353118896484375,
-0.04388427734375,
0.01212310791015625,
-0.0160064697265625,
-0.0465087890625,
0.0019779205322265625,
-0.01464080810546875,
-0.00014591217041015625,
-0.001087188720703125,
0.04547119140625,
-0.006561279296875,
-0.03948974609375,
0.0263824462890625,
0.0096893310546875,
0.0191192626953125,
0.01393890380859375,
-0.06781005859375,
0.0217742919921875,
-0.00213623046875,
-0.0340576171875,
0.031280517578125,
0.0206451416015625,
0.01222991943359375,
0.04998779296875,
0.0587158203125,
-0.0134124755859375,
0.005275726318359375,
0.0008130073547363281,
0.08050537109375,
-0.0552978515625,
-0.031494140625,
-0.06829833984375,
0.0455322265625,
0.0023975372314453125,
-0.037322998046875,
0.05487060546875,
0.04327392578125,
0.06500244140625,
-0.00867462158203125,
0.0467529296875,
-0.0263824462890625,
-0.004817962646484375,
-0.0208282470703125,
0.060028076171875,
-0.0546875,
0.0164031982421875,
-0.0201416015625,
-0.054779052734375,
-0.0249176025390625,
0.05413818359375,
-0.01052093505859375,
0.0038509368896484375,
0.0302581787109375,
0.08013916015625,
-0.00955963134765625,
0.00002390146255493164,
0.0252532958984375,
0.03717041015625,
0.026458740234375,
0.0653076171875,
0.059600830078125,
-0.0731201171875,
0.0445556640625,
-0.035614013671875,
-0.0199737548828125,
-0.0202484130859375,
-0.063232421875,
-0.046234130859375,
-0.020172119140625,
-0.038360595703125,
-0.037139892578125,
0.002017974853515625,
0.0703125,
0.0706787109375,
-0.057952880859375,
-0.0243377685546875,
-0.0151824951171875,
0.0008187294006347656,
-0.013763427734375,
-0.0199737548828125,
0.0231475830078125,
0.016021728515625,
-0.05316162109375,
0.01959228515625,
0.01171112060546875,
0.031280517578125,
-0.012542724609375,
-0.01134490966796875,
-0.03289794921875,
0.0008006095886230469,
0.0311126708984375,
0.0465087890625,
-0.0550537109375,
-0.0202789306640625,
0.0008473396301269531,
-0.0015468597412109375,
0.01499176025390625,
0.0304718017578125,
-0.053466796875,
0.01253509521484375,
0.033203125,
0.027435302734375,
0.043701171875,
-0.0081787109375,
0.0286407470703125,
-0.0310821533203125,
0.0200653076171875,
0.0015010833740234375,
0.0330810546875,
0.017730712890625,
-0.039947509765625,
0.039031982421875,
0.029571533203125,
-0.050262451171875,
-0.068359375,
-0.007152557373046875,
-0.0733642578125,
-0.0199432373046875,
0.0758056640625,
-0.0246429443359375,
-0.04010009765625,
0.017242431640625,
-0.0179595947265625,
0.037109375,
-0.0259246826171875,
0.0400390625,
0.03057861328125,
-0.024017333984375,
-0.0181121826171875,
-0.036224365234375,
0.0209808349609375,
0.033416748046875,
-0.052886962890625,
0.0035552978515625,
0.03546142578125,
0.0311737060546875,
0.0250091552734375,
0.057525634765625,
-0.009674072265625,
0.039031982421875,
0.020477294921875,
0.00792694091796875,
-0.00445556640625,
-0.0212554931640625,
-0.029754638671875,
-0.0025634765625,
-0.017059326171875,
-0.0159149169921875
]
] |
timm/resnet34.tv_in1k | 2023-04-05T18:07:29.000Z | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"arxiv:1512.03385",
"license:bsd-3-clause",
"region:us"
] | image-classification | timm | null | null | timm/resnet34.tv_in1k | 0 | 7,375 | timm | 2023-04-05T18:07:16 | ---
tags:
- image-classification
- timm
library_tag: timm
license: bsd-3-clause
---
# Model card for resnet34.tv_in1k
A ResNet-B image classification model.
This model features:
* ReLU activations
* single layer 7x7 convolution with pooling
* 1x1 convolution shortcut downsample
Trained on ImageNet-1k, original torchvision model weight.
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 21.8
- GMACs: 3.7
- Activations (M): 3.7
- Image size: 224 x 224
- **Papers:**
- Deep Residual Learning for Image Recognition: https://arxiv.org/abs/1512.03385
- **Original:** https://github.com/pytorch/vision
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('resnet34.tv_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Feature Map Extraction
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'resnet34.tv_in1k',
pretrained=True,
features_only=True,
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
for o in output:
# print shape of each feature map in output
# e.g.:
# torch.Size([1, 64, 112, 112])
# torch.Size([1, 64, 56, 56])
# torch.Size([1, 128, 28, 28])
# torch.Size([1, 256, 14, 14])
# torch.Size([1, 512, 7, 7])
print(o.shape)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'resnet34.tv_in1k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 512, 7, 7) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
|model |img_size|top1 |top5 |param_count|gmacs|macts|img/sec|
|------------------------------------------|--------|-----|-----|-----------|-----|-----|-------|
|[seresnextaa101d_32x8d.sw_in12k_ft_in1k_288](https://huggingface.co/timm/seresnextaa101d_32x8d.sw_in12k_ft_in1k_288)|320 |86.72|98.17|93.6 |35.2 |69.7 |451 |
|[seresnextaa101d_32x8d.sw_in12k_ft_in1k_288](https://huggingface.co/timm/seresnextaa101d_32x8d.sw_in12k_ft_in1k_288)|288 |86.51|98.08|93.6 |28.5 |56.4 |560 |
|[seresnextaa101d_32x8d.sw_in12k_ft_in1k](https://huggingface.co/timm/seresnextaa101d_32x8d.sw_in12k_ft_in1k)|288 |86.49|98.03|93.6 |28.5 |56.4 |557 |
|[seresnextaa101d_32x8d.sw_in12k_ft_in1k](https://huggingface.co/timm/seresnextaa101d_32x8d.sw_in12k_ft_in1k)|224 |85.96|97.82|93.6 |17.2 |34.2 |923 |
|[resnext101_32x32d.fb_wsl_ig1b_ft_in1k](https://huggingface.co/timm/resnext101_32x32d.fb_wsl_ig1b_ft_in1k)|224 |85.11|97.44|468.5 |87.3 |91.1 |254 |
|[resnetrs420.tf_in1k](https://huggingface.co/timm/resnetrs420.tf_in1k)|416 |85.0 |97.12|191.9 |108.4|213.8|134 |
|[ecaresnet269d.ra2_in1k](https://huggingface.co/timm/ecaresnet269d.ra2_in1k)|352 |84.96|97.22|102.1 |50.2 |101.2|291 |
|[ecaresnet269d.ra2_in1k](https://huggingface.co/timm/ecaresnet269d.ra2_in1k)|320 |84.73|97.18|102.1 |41.5 |83.7 |353 |
|[resnetrs350.tf_in1k](https://huggingface.co/timm/resnetrs350.tf_in1k)|384 |84.71|96.99|164.0 |77.6 |154.7|183 |
|[seresnextaa101d_32x8d.ah_in1k](https://huggingface.co/timm/seresnextaa101d_32x8d.ah_in1k)|288 |84.57|97.08|93.6 |28.5 |56.4 |557 |
|[resnetrs200.tf_in1k](https://huggingface.co/timm/resnetrs200.tf_in1k)|320 |84.45|97.08|93.2 |31.5 |67.8 |446 |
|[resnetrs270.tf_in1k](https://huggingface.co/timm/resnetrs270.tf_in1k)|352 |84.43|96.97|129.9 |51.1 |105.5|280 |
|[seresnext101d_32x8d.ah_in1k](https://huggingface.co/timm/seresnext101d_32x8d.ah_in1k)|288 |84.36|96.92|93.6 |27.6 |53.0 |595 |
|[seresnet152d.ra2_in1k](https://huggingface.co/timm/seresnet152d.ra2_in1k)|320 |84.35|97.04|66.8 |24.1 |47.7 |610 |
|[resnetrs350.tf_in1k](https://huggingface.co/timm/resnetrs350.tf_in1k)|288 |84.3 |96.94|164.0 |43.7 |87.1 |333 |
|[resnext101_32x8d.fb_swsl_ig1b_ft_in1k](https://huggingface.co/timm/resnext101_32x8d.fb_swsl_ig1b_ft_in1k)|224 |84.28|97.17|88.8 |16.5 |31.2 |1100 |
|[resnetrs420.tf_in1k](https://huggingface.co/timm/resnetrs420.tf_in1k)|320 |84.24|96.86|191.9 |64.2 |126.6|228 |
|[seresnext101_32x8d.ah_in1k](https://huggingface.co/timm/seresnext101_32x8d.ah_in1k)|288 |84.19|96.87|93.6 |27.2 |51.6 |613 |
|[resnext101_32x16d.fb_wsl_ig1b_ft_in1k](https://huggingface.co/timm/resnext101_32x16d.fb_wsl_ig1b_ft_in1k)|224 |84.18|97.19|194.0 |36.3 |51.2 |581 |
|[resnetaa101d.sw_in12k_ft_in1k](https://huggingface.co/timm/resnetaa101d.sw_in12k_ft_in1k)|288 |84.11|97.11|44.6 |15.1 |29.0 |1144 |
|[resnet200d.ra2_in1k](https://huggingface.co/timm/resnet200d.ra2_in1k)|320 |83.97|96.82|64.7 |31.2 |67.3 |518 |
|[resnetrs200.tf_in1k](https://huggingface.co/timm/resnetrs200.tf_in1k)|256 |83.87|96.75|93.2 |20.2 |43.4 |692 |
|[seresnextaa101d_32x8d.ah_in1k](https://huggingface.co/timm/seresnextaa101d_32x8d.ah_in1k)|224 |83.86|96.65|93.6 |17.2 |34.2 |923 |
|[resnetrs152.tf_in1k](https://huggingface.co/timm/resnetrs152.tf_in1k)|320 |83.72|96.61|86.6 |24.3 |48.1 |617 |
|[seresnet152d.ra2_in1k](https://huggingface.co/timm/seresnet152d.ra2_in1k)|256 |83.69|96.78|66.8 |15.4 |30.6 |943 |
|[seresnext101d_32x8d.ah_in1k](https://huggingface.co/timm/seresnext101d_32x8d.ah_in1k)|224 |83.68|96.61|93.6 |16.7 |32.0 |986 |
|[resnet152d.ra2_in1k](https://huggingface.co/timm/resnet152d.ra2_in1k)|320 |83.67|96.74|60.2 |24.1 |47.7 |706 |
|[resnetrs270.tf_in1k](https://huggingface.co/timm/resnetrs270.tf_in1k)|256 |83.59|96.61|129.9 |27.1 |55.8 |526 |
|[seresnext101_32x8d.ah_in1k](https://huggingface.co/timm/seresnext101_32x8d.ah_in1k)|224 |83.58|96.4 |93.6 |16.5 |31.2 |1013 |
|[resnetaa101d.sw_in12k_ft_in1k](https://huggingface.co/timm/resnetaa101d.sw_in12k_ft_in1k)|224 |83.54|96.83|44.6 |9.1 |17.6 |1864 |
|[resnet152.a1h_in1k](https://huggingface.co/timm/resnet152.a1h_in1k)|288 |83.46|96.54|60.2 |19.1 |37.3 |904 |
|[resnext101_32x16d.fb_swsl_ig1b_ft_in1k](https://huggingface.co/timm/resnext101_32x16d.fb_swsl_ig1b_ft_in1k)|224 |83.35|96.85|194.0 |36.3 |51.2 |582 |
|[resnet200d.ra2_in1k](https://huggingface.co/timm/resnet200d.ra2_in1k)|256 |83.23|96.53|64.7 |20.0 |43.1 |809 |
|[resnext101_32x4d.fb_swsl_ig1b_ft_in1k](https://huggingface.co/timm/resnext101_32x4d.fb_swsl_ig1b_ft_in1k)|224 |83.22|96.75|44.2 |8.0 |21.2 |1814 |
|[resnext101_64x4d.c1_in1k](https://huggingface.co/timm/resnext101_64x4d.c1_in1k)|288 |83.16|96.38|83.5 |25.7 |51.6 |590 |
|[resnet152d.ra2_in1k](https://huggingface.co/timm/resnet152d.ra2_in1k)|256 |83.14|96.38|60.2 |15.4 |30.5 |1096 |
|[resnet101d.ra2_in1k](https://huggingface.co/timm/resnet101d.ra2_in1k)|320 |83.02|96.45|44.6 |16.5 |34.8 |992 |
|[ecaresnet101d.miil_in1k](https://huggingface.co/timm/ecaresnet101d.miil_in1k)|288 |82.98|96.54|44.6 |13.4 |28.2 |1077 |
|[resnext101_64x4d.tv_in1k](https://huggingface.co/timm/resnext101_64x4d.tv_in1k)|224 |82.98|96.25|83.5 |15.5 |31.2 |989 |
|[resnetrs152.tf_in1k](https://huggingface.co/timm/resnetrs152.tf_in1k)|256 |82.86|96.28|86.6 |15.6 |30.8 |951 |
|[resnext101_32x8d.tv2_in1k](https://huggingface.co/timm/resnext101_32x8d.tv2_in1k)|224 |82.83|96.22|88.8 |16.5 |31.2 |1099 |
|[resnet152.a1h_in1k](https://huggingface.co/timm/resnet152.a1h_in1k)|224 |82.8 |96.13|60.2 |11.6 |22.6 |1486 |
|[resnet101.a1h_in1k](https://huggingface.co/timm/resnet101.a1h_in1k)|288 |82.8 |96.32|44.6 |13.0 |26.8 |1291 |
|[resnet152.a1_in1k](https://huggingface.co/timm/resnet152.a1_in1k)|288 |82.74|95.71|60.2 |19.1 |37.3 |905 |
|[resnext101_32x8d.fb_wsl_ig1b_ft_in1k](https://huggingface.co/timm/resnext101_32x8d.fb_wsl_ig1b_ft_in1k)|224 |82.69|96.63|88.8 |16.5 |31.2 |1100 |
|[resnet152.a2_in1k](https://huggingface.co/timm/resnet152.a2_in1k)|288 |82.62|95.75|60.2 |19.1 |37.3 |904 |
|[resnetaa50d.sw_in12k_ft_in1k](https://huggingface.co/timm/resnetaa50d.sw_in12k_ft_in1k)|288 |82.61|96.49|25.6 |8.9 |20.6 |1729 |
|[resnet61q.ra2_in1k](https://huggingface.co/timm/resnet61q.ra2_in1k)|288 |82.53|96.13|36.8 |9.9 |21.5 |1773 |
|[wide_resnet101_2.tv2_in1k](https://huggingface.co/timm/wide_resnet101_2.tv2_in1k)|224 |82.5 |96.02|126.9 |22.8 |21.2 |1078 |
|[resnext101_64x4d.c1_in1k](https://huggingface.co/timm/resnext101_64x4d.c1_in1k)|224 |82.46|95.92|83.5 |15.5 |31.2 |987 |
|[resnet51q.ra2_in1k](https://huggingface.co/timm/resnet51q.ra2_in1k)|288 |82.36|96.18|35.7 |8.1 |20.9 |1964 |
|[ecaresnet50t.ra2_in1k](https://huggingface.co/timm/ecaresnet50t.ra2_in1k)|320 |82.35|96.14|25.6 |8.8 |24.1 |1386 |
|[resnet101.a1_in1k](https://huggingface.co/timm/resnet101.a1_in1k)|288 |82.31|95.63|44.6 |13.0 |26.8 |1291 |
|[resnetrs101.tf_in1k](https://huggingface.co/timm/resnetrs101.tf_in1k)|288 |82.29|96.01|63.6 |13.6 |28.5 |1078 |
|[resnet152.tv2_in1k](https://huggingface.co/timm/resnet152.tv2_in1k)|224 |82.29|96.0 |60.2 |11.6 |22.6 |1484 |
|[wide_resnet50_2.racm_in1k](https://huggingface.co/timm/wide_resnet50_2.racm_in1k)|288 |82.27|96.06|68.9 |18.9 |23.8 |1176 |
|[resnet101d.ra2_in1k](https://huggingface.co/timm/resnet101d.ra2_in1k)|256 |82.26|96.07|44.6 |10.6 |22.2 |1542 |
|[resnet101.a2_in1k](https://huggingface.co/timm/resnet101.a2_in1k)|288 |82.24|95.73|44.6 |13.0 |26.8 |1290 |
|[seresnext50_32x4d.racm_in1k](https://huggingface.co/timm/seresnext50_32x4d.racm_in1k)|288 |82.2 |96.14|27.6 |7.0 |23.8 |1547 |
|[ecaresnet101d.miil_in1k](https://huggingface.co/timm/ecaresnet101d.miil_in1k)|224 |82.18|96.05|44.6 |8.1 |17.1 |1771 |
|[resnext50_32x4d.fb_swsl_ig1b_ft_in1k](https://huggingface.co/timm/resnext50_32x4d.fb_swsl_ig1b_ft_in1k)|224 |82.17|96.22|25.0 |4.3 |14.4 |2943 |
|[ecaresnet50t.a1_in1k](https://huggingface.co/timm/ecaresnet50t.a1_in1k)|288 |82.12|95.65|25.6 |7.1 |19.6 |1704 |
|[resnext50_32x4d.a1h_in1k](https://huggingface.co/timm/resnext50_32x4d.a1h_in1k)|288 |82.03|95.94|25.0 |7.0 |23.8 |1745 |
|[ecaresnet101d_pruned.miil_in1k](https://huggingface.co/timm/ecaresnet101d_pruned.miil_in1k)|288 |82.0 |96.15|24.9 |5.8 |12.7 |1787 |
|[resnet61q.ra2_in1k](https://huggingface.co/timm/resnet61q.ra2_in1k)|256 |81.99|95.85|36.8 |7.8 |17.0 |2230 |
|[resnext101_32x8d.tv2_in1k](https://huggingface.co/timm/resnext101_32x8d.tv2_in1k)|176 |81.98|95.72|88.8 |10.3 |19.4 |1768 |
|[resnet152.a1_in1k](https://huggingface.co/timm/resnet152.a1_in1k)|224 |81.97|95.24|60.2 |11.6 |22.6 |1486 |
|[resnet101.a1h_in1k](https://huggingface.co/timm/resnet101.a1h_in1k)|224 |81.93|95.75|44.6 |7.8 |16.2 |2122 |
|[resnet101.tv2_in1k](https://huggingface.co/timm/resnet101.tv2_in1k)|224 |81.9 |95.77|44.6 |7.8 |16.2 |2118 |
|[resnext101_32x16d.fb_ssl_yfcc100m_ft_in1k](https://huggingface.co/timm/resnext101_32x16d.fb_ssl_yfcc100m_ft_in1k)|224 |81.84|96.1 |194.0 |36.3 |51.2 |583 |
|[resnet51q.ra2_in1k](https://huggingface.co/timm/resnet51q.ra2_in1k)|256 |81.78|95.94|35.7 |6.4 |16.6 |2471 |
|[resnet152.a2_in1k](https://huggingface.co/timm/resnet152.a2_in1k)|224 |81.77|95.22|60.2 |11.6 |22.6 |1485 |
|[resnetaa50d.sw_in12k_ft_in1k](https://huggingface.co/timm/resnetaa50d.sw_in12k_ft_in1k)|224 |81.74|96.06|25.6 |5.4 |12.4 |2813 |
|[ecaresnet50t.a2_in1k](https://huggingface.co/timm/ecaresnet50t.a2_in1k)|288 |81.65|95.54|25.6 |7.1 |19.6 |1703 |
|[ecaresnet50d.miil_in1k](https://huggingface.co/timm/ecaresnet50d.miil_in1k)|288 |81.64|95.88|25.6 |7.2 |19.7 |1694 |
|[resnext101_32x8d.fb_ssl_yfcc100m_ft_in1k](https://huggingface.co/timm/resnext101_32x8d.fb_ssl_yfcc100m_ft_in1k)|224 |81.62|96.04|88.8 |16.5 |31.2 |1101 |
|[wide_resnet50_2.tv2_in1k](https://huggingface.co/timm/wide_resnet50_2.tv2_in1k)|224 |81.61|95.76|68.9 |11.4 |14.4 |1930 |
|[resnetaa50.a1h_in1k](https://huggingface.co/timm/resnetaa50.a1h_in1k)|288 |81.61|95.83|25.6 |8.5 |19.2 |1868 |
|[resnet101.a1_in1k](https://huggingface.co/timm/resnet101.a1_in1k)|224 |81.5 |95.16|44.6 |7.8 |16.2 |2125 |
|[resnext50_32x4d.a1_in1k](https://huggingface.co/timm/resnext50_32x4d.a1_in1k)|288 |81.48|95.16|25.0 |7.0 |23.8 |1745 |
|[gcresnet50t.ra2_in1k](https://huggingface.co/timm/gcresnet50t.ra2_in1k)|288 |81.47|95.71|25.9 |6.9 |18.6 |2071 |
|[wide_resnet50_2.racm_in1k](https://huggingface.co/timm/wide_resnet50_2.racm_in1k)|224 |81.45|95.53|68.9 |11.4 |14.4 |1929 |
|[resnet50d.a1_in1k](https://huggingface.co/timm/resnet50d.a1_in1k)|288 |81.44|95.22|25.6 |7.2 |19.7 |1908 |
|[ecaresnet50t.ra2_in1k](https://huggingface.co/timm/ecaresnet50t.ra2_in1k)|256 |81.44|95.67|25.6 |5.6 |15.4 |2168 |
|[ecaresnetlight.miil_in1k](https://huggingface.co/timm/ecaresnetlight.miil_in1k)|288 |81.4 |95.82|30.2 |6.8 |13.9 |2132 |
|[resnet50d.ra2_in1k](https://huggingface.co/timm/resnet50d.ra2_in1k)|288 |81.37|95.74|25.6 |7.2 |19.7 |1910 |
|[resnet101.a2_in1k](https://huggingface.co/timm/resnet101.a2_in1k)|224 |81.32|95.19|44.6 |7.8 |16.2 |2125 |
|[seresnet50.ra2_in1k](https://huggingface.co/timm/seresnet50.ra2_in1k)|288 |81.3 |95.65|28.1 |6.8 |18.4 |1803 |
|[resnext50_32x4d.a2_in1k](https://huggingface.co/timm/resnext50_32x4d.a2_in1k)|288 |81.3 |95.11|25.0 |7.0 |23.8 |1746 |
|[seresnext50_32x4d.racm_in1k](https://huggingface.co/timm/seresnext50_32x4d.racm_in1k)|224 |81.27|95.62|27.6 |4.3 |14.4 |2591 |
|[ecaresnet50t.a1_in1k](https://huggingface.co/timm/ecaresnet50t.a1_in1k)|224 |81.26|95.16|25.6 |4.3 |11.8 |2823 |
|[gcresnext50ts.ch_in1k](https://huggingface.co/timm/gcresnext50ts.ch_in1k)|288 |81.23|95.54|15.7 |4.8 |19.6 |2117 |
|[senet154.gluon_in1k](https://huggingface.co/timm/senet154.gluon_in1k)|224 |81.23|95.35|115.1 |20.8 |38.7 |545 |
|[resnet50.a1_in1k](https://huggingface.co/timm/resnet50.a1_in1k)|288 |81.22|95.11|25.6 |6.8 |18.4 |2089 |
|[resnet50_gn.a1h_in1k](https://huggingface.co/timm/resnet50_gn.a1h_in1k)|288 |81.22|95.63|25.6 |6.8 |18.4 |676 |
|[resnet50d.a2_in1k](https://huggingface.co/timm/resnet50d.a2_in1k)|288 |81.18|95.09|25.6 |7.2 |19.7 |1908 |
|[resnet50.fb_swsl_ig1b_ft_in1k](https://huggingface.co/timm/resnet50.fb_swsl_ig1b_ft_in1k)|224 |81.18|95.98|25.6 |4.1 |11.1 |3455 |
|[resnext50_32x4d.tv2_in1k](https://huggingface.co/timm/resnext50_32x4d.tv2_in1k)|224 |81.17|95.34|25.0 |4.3 |14.4 |2933 |
|[resnext50_32x4d.a1h_in1k](https://huggingface.co/timm/resnext50_32x4d.a1h_in1k)|224 |81.1 |95.33|25.0 |4.3 |14.4 |2934 |
|[seresnet50.a2_in1k](https://huggingface.co/timm/seresnet50.a2_in1k)|288 |81.1 |95.23|28.1 |6.8 |18.4 |1801 |
|[seresnet50.a1_in1k](https://huggingface.co/timm/seresnet50.a1_in1k)|288 |81.1 |95.12|28.1 |6.8 |18.4 |1799 |
|[resnet152s.gluon_in1k](https://huggingface.co/timm/resnet152s.gluon_in1k)|224 |81.02|95.41|60.3 |12.9 |25.0 |1347 |
|[resnet50.d_in1k](https://huggingface.co/timm/resnet50.d_in1k)|288 |80.97|95.44|25.6 |6.8 |18.4 |2085 |
|[gcresnet50t.ra2_in1k](https://huggingface.co/timm/gcresnet50t.ra2_in1k)|256 |80.94|95.45|25.9 |5.4 |14.7 |2571 |
|[resnext101_32x4d.fb_ssl_yfcc100m_ft_in1k](https://huggingface.co/timm/resnext101_32x4d.fb_ssl_yfcc100m_ft_in1k)|224 |80.93|95.73|44.2 |8.0 |21.2 |1814 |
|[resnet50.c1_in1k](https://huggingface.co/timm/resnet50.c1_in1k)|288 |80.91|95.55|25.6 |6.8 |18.4 |2084 |
|[seresnext101_32x4d.gluon_in1k](https://huggingface.co/timm/seresnext101_32x4d.gluon_in1k)|224 |80.9 |95.31|49.0 |8.0 |21.3 |1585 |
|[seresnext101_64x4d.gluon_in1k](https://huggingface.co/timm/seresnext101_64x4d.gluon_in1k)|224 |80.9 |95.3 |88.2 |15.5 |31.2 |918 |
|[resnet50.c2_in1k](https://huggingface.co/timm/resnet50.c2_in1k)|288 |80.86|95.52|25.6 |6.8 |18.4 |2085 |
|[resnet50.tv2_in1k](https://huggingface.co/timm/resnet50.tv2_in1k)|224 |80.85|95.43|25.6 |4.1 |11.1 |3450 |
|[ecaresnet50t.a2_in1k](https://huggingface.co/timm/ecaresnet50t.a2_in1k)|224 |80.84|95.02|25.6 |4.3 |11.8 |2821 |
|[ecaresnet101d_pruned.miil_in1k](https://huggingface.co/timm/ecaresnet101d_pruned.miil_in1k)|224 |80.79|95.62|24.9 |3.5 |7.7 |2961 |
|[seresnet33ts.ra2_in1k](https://huggingface.co/timm/seresnet33ts.ra2_in1k)|288 |80.79|95.36|19.8 |6.0 |14.8 |2506 |
|[ecaresnet50d_pruned.miil_in1k](https://huggingface.co/timm/ecaresnet50d_pruned.miil_in1k)|288 |80.79|95.58|19.9 |4.2 |10.6 |2349 |
|[resnet50.a2_in1k](https://huggingface.co/timm/resnet50.a2_in1k)|288 |80.78|94.99|25.6 |6.8 |18.4 |2088 |
|[resnet50.b1k_in1k](https://huggingface.co/timm/resnet50.b1k_in1k)|288 |80.71|95.43|25.6 |6.8 |18.4 |2087 |
|[resnext50_32x4d.ra_in1k](https://huggingface.co/timm/resnext50_32x4d.ra_in1k)|288 |80.7 |95.39|25.0 |7.0 |23.8 |1749 |
|[resnetrs101.tf_in1k](https://huggingface.co/timm/resnetrs101.tf_in1k)|192 |80.69|95.24|63.6 |6.0 |12.7 |2270 |
|[resnet50d.a1_in1k](https://huggingface.co/timm/resnet50d.a1_in1k)|224 |80.68|94.71|25.6 |4.4 |11.9 |3162 |
|[eca_resnet33ts.ra2_in1k](https://huggingface.co/timm/eca_resnet33ts.ra2_in1k)|288 |80.68|95.36|19.7 |6.0 |14.8 |2637 |
|[resnet50.a1h_in1k](https://huggingface.co/timm/resnet50.a1h_in1k)|224 |80.67|95.3 |25.6 |4.1 |11.1 |3452 |
|[resnext50d_32x4d.bt_in1k](https://huggingface.co/timm/resnext50d_32x4d.bt_in1k)|288 |80.67|95.42|25.0 |7.4 |25.1 |1626 |
|[resnetaa50.a1h_in1k](https://huggingface.co/timm/resnetaa50.a1h_in1k)|224 |80.63|95.21|25.6 |5.2 |11.6 |3034 |
|[ecaresnet50d.miil_in1k](https://huggingface.co/timm/ecaresnet50d.miil_in1k)|224 |80.61|95.32|25.6 |4.4 |11.9 |2813 |
|[resnext101_64x4d.gluon_in1k](https://huggingface.co/timm/resnext101_64x4d.gluon_in1k)|224 |80.61|94.99|83.5 |15.5 |31.2 |989 |
|[gcresnet33ts.ra2_in1k](https://huggingface.co/timm/gcresnet33ts.ra2_in1k)|288 |80.6 |95.31|19.9 |6.0 |14.8 |2578 |
|[gcresnext50ts.ch_in1k](https://huggingface.co/timm/gcresnext50ts.ch_in1k)|256 |80.57|95.17|15.7 |3.8 |15.5 |2710 |
|[resnet152.a3_in1k](https://huggingface.co/timm/resnet152.a3_in1k)|224 |80.56|95.0 |60.2 |11.6 |22.6 |1483 |
|[resnet50d.ra2_in1k](https://huggingface.co/timm/resnet50d.ra2_in1k)|224 |80.53|95.16|25.6 |4.4 |11.9 |3164 |
|[resnext50_32x4d.a1_in1k](https://huggingface.co/timm/resnext50_32x4d.a1_in1k)|224 |80.53|94.46|25.0 |4.3 |14.4 |2930 |
|[wide_resnet101_2.tv2_in1k](https://huggingface.co/timm/wide_resnet101_2.tv2_in1k)|176 |80.48|94.98|126.9 |14.3 |13.2 |1719 |
|[resnet152d.gluon_in1k](https://huggingface.co/timm/resnet152d.gluon_in1k)|224 |80.47|95.2 |60.2 |11.8 |23.4 |1428 |
|[resnet50.b2k_in1k](https://huggingface.co/timm/resnet50.b2k_in1k)|288 |80.45|95.32|25.6 |6.8 |18.4 |2086 |
|[ecaresnetlight.miil_in1k](https://huggingface.co/timm/ecaresnetlight.miil_in1k)|224 |80.45|95.24|30.2 |4.1 |8.4 |3530 |
|[resnext50_32x4d.a2_in1k](https://huggingface.co/timm/resnext50_32x4d.a2_in1k)|224 |80.45|94.63|25.0 |4.3 |14.4 |2936 |
|[wide_resnet50_2.tv2_in1k](https://huggingface.co/timm/wide_resnet50_2.tv2_in1k)|176 |80.43|95.09|68.9 |7.3 |9.0 |3015 |
|[resnet101d.gluon_in1k](https://huggingface.co/timm/resnet101d.gluon_in1k)|224 |80.42|95.01|44.6 |8.1 |17.0 |2007 |
|[resnet50.a1_in1k](https://huggingface.co/timm/resnet50.a1_in1k)|224 |80.38|94.6 |25.6 |4.1 |11.1 |3461 |
|[seresnet33ts.ra2_in1k](https://huggingface.co/timm/seresnet33ts.ra2_in1k)|256 |80.36|95.1 |19.8 |4.8 |11.7 |3267 |
|[resnext101_32x4d.gluon_in1k](https://huggingface.co/timm/resnext101_32x4d.gluon_in1k)|224 |80.34|94.93|44.2 |8.0 |21.2 |1814 |
|[resnext50_32x4d.fb_ssl_yfcc100m_ft_in1k](https://huggingface.co/timm/resnext50_32x4d.fb_ssl_yfcc100m_ft_in1k)|224 |80.32|95.4 |25.0 |4.3 |14.4 |2941 |
|[resnet101s.gluon_in1k](https://huggingface.co/timm/resnet101s.gluon_in1k)|224 |80.28|95.16|44.7 |9.2 |18.6 |1851 |
|[seresnet50.ra2_in1k](https://huggingface.co/timm/seresnet50.ra2_in1k)|224 |80.26|95.08|28.1 |4.1 |11.1 |2972 |
|[resnetblur50.bt_in1k](https://huggingface.co/timm/resnetblur50.bt_in1k)|288 |80.24|95.24|25.6 |8.5 |19.9 |1523 |
|[resnet50d.a2_in1k](https://huggingface.co/timm/resnet50d.a2_in1k)|224 |80.22|94.63|25.6 |4.4 |11.9 |3162 |
|[resnet152.tv2_in1k](https://huggingface.co/timm/resnet152.tv2_in1k)|176 |80.2 |94.64|60.2 |7.2 |14.0 |2346 |
|[seresnet50.a2_in1k](https://huggingface.co/timm/seresnet50.a2_in1k)|224 |80.08|94.74|28.1 |4.1 |11.1 |2969 |
|[eca_resnet33ts.ra2_in1k](https://huggingface.co/timm/eca_resnet33ts.ra2_in1k)|256 |80.08|94.97|19.7 |4.8 |11.7 |3284 |
|[gcresnet33ts.ra2_in1k](https://huggingface.co/timm/gcresnet33ts.ra2_in1k)|256 |80.06|94.99|19.9 |4.8 |11.7 |3216 |
|[resnet50_gn.a1h_in1k](https://huggingface.co/timm/resnet50_gn.a1h_in1k)|224 |80.06|94.95|25.6 |4.1 |11.1 |1109 |
|[seresnet50.a1_in1k](https://huggingface.co/timm/seresnet50.a1_in1k)|224 |80.02|94.71|28.1 |4.1 |11.1 |2962 |
|[resnet50.ram_in1k](https://huggingface.co/timm/resnet50.ram_in1k)|288 |79.97|95.05|25.6 |6.8 |18.4 |2086 |
|[resnet152c.gluon_in1k](https://huggingface.co/timm/resnet152c.gluon_in1k)|224 |79.92|94.84|60.2 |11.8 |23.4 |1455 |
|[seresnext50_32x4d.gluon_in1k](https://huggingface.co/timm/seresnext50_32x4d.gluon_in1k)|224 |79.91|94.82|27.6 |4.3 |14.4 |2591 |
|[resnet50.d_in1k](https://huggingface.co/timm/resnet50.d_in1k)|224 |79.91|94.67|25.6 |4.1 |11.1 |3456 |
|[resnet101.tv2_in1k](https://huggingface.co/timm/resnet101.tv2_in1k)|176 |79.9 |94.6 |44.6 |4.9 |10.1 |3341 |
|[resnetrs50.tf_in1k](https://huggingface.co/timm/resnetrs50.tf_in1k)|224 |79.89|94.97|35.7 |4.5 |12.1 |2774 |
|[resnet50.c2_in1k](https://huggingface.co/timm/resnet50.c2_in1k)|224 |79.88|94.87|25.6 |4.1 |11.1 |3455 |
|[ecaresnet26t.ra2_in1k](https://huggingface.co/timm/ecaresnet26t.ra2_in1k)|320 |79.86|95.07|16.0 |5.2 |16.4 |2168 |
|[resnet50.a2_in1k](https://huggingface.co/timm/resnet50.a2_in1k)|224 |79.85|94.56|25.6 |4.1 |11.1 |3460 |
|[resnet50.ra_in1k](https://huggingface.co/timm/resnet50.ra_in1k)|288 |79.83|94.97|25.6 |6.8 |18.4 |2087 |
|[resnet101.a3_in1k](https://huggingface.co/timm/resnet101.a3_in1k)|224 |79.82|94.62|44.6 |7.8 |16.2 |2114 |
|[resnext50_32x4d.ra_in1k](https://huggingface.co/timm/resnext50_32x4d.ra_in1k)|224 |79.76|94.6 |25.0 |4.3 |14.4 |2943 |
|[resnet50.c1_in1k](https://huggingface.co/timm/resnet50.c1_in1k)|224 |79.74|94.95|25.6 |4.1 |11.1 |3455 |
|[ecaresnet50d_pruned.miil_in1k](https://huggingface.co/timm/ecaresnet50d_pruned.miil_in1k)|224 |79.74|94.87|19.9 |2.5 |6.4 |3929 |
|[resnet33ts.ra2_in1k](https://huggingface.co/timm/resnet33ts.ra2_in1k)|288 |79.71|94.83|19.7 |6.0 |14.8 |2710 |
|[resnet152.gluon_in1k](https://huggingface.co/timm/resnet152.gluon_in1k)|224 |79.68|94.74|60.2 |11.6 |22.6 |1486 |
|[resnext50d_32x4d.bt_in1k](https://huggingface.co/timm/resnext50d_32x4d.bt_in1k)|224 |79.67|94.87|25.0 |4.5 |15.2 |2729 |
|[resnet50.bt_in1k](https://huggingface.co/timm/resnet50.bt_in1k)|288 |79.63|94.91|25.6 |6.8 |18.4 |2086 |
|[ecaresnet50t.a3_in1k](https://huggingface.co/timm/ecaresnet50t.a3_in1k)|224 |79.56|94.72|25.6 |4.3 |11.8 |2805 |
|[resnet101c.gluon_in1k](https://huggingface.co/timm/resnet101c.gluon_in1k)|224 |79.53|94.58|44.6 |8.1 |17.0 |2062 |
|[resnet50.b1k_in1k](https://huggingface.co/timm/resnet50.b1k_in1k)|224 |79.52|94.61|25.6 |4.1 |11.1 |3459 |
|[resnet50.tv2_in1k](https://huggingface.co/timm/resnet50.tv2_in1k)|176 |79.42|94.64|25.6 |2.6 |6.9 |5397 |
|[resnet32ts.ra2_in1k](https://huggingface.co/timm/resnet32ts.ra2_in1k)|288 |79.4 |94.66|18.0 |5.9 |14.6 |2752 |
|[resnet50.b2k_in1k](https://huggingface.co/timm/resnet50.b2k_in1k)|224 |79.38|94.57|25.6 |4.1 |11.1 |3459 |
|[resnext50_32x4d.tv2_in1k](https://huggingface.co/timm/resnext50_32x4d.tv2_in1k)|176 |79.37|94.3 |25.0 |2.7 |9.0 |4577 |
|[resnext50_32x4d.gluon_in1k](https://huggingface.co/timm/resnext50_32x4d.gluon_in1k)|224 |79.36|94.43|25.0 |4.3 |14.4 |2942 |
|[resnext101_32x8d.tv_in1k](https://huggingface.co/timm/resnext101_32x8d.tv_in1k)|224 |79.31|94.52|88.8 |16.5 |31.2 |1100 |
|[resnet101.gluon_in1k](https://huggingface.co/timm/resnet101.gluon_in1k)|224 |79.31|94.53|44.6 |7.8 |16.2 |2125 |
|[resnetblur50.bt_in1k](https://huggingface.co/timm/resnetblur50.bt_in1k)|224 |79.31|94.63|25.6 |5.2 |12.0 |2524 |
|[resnet50.a1h_in1k](https://huggingface.co/timm/resnet50.a1h_in1k)|176 |79.27|94.49|25.6 |2.6 |6.9 |5404 |
|[resnext50_32x4d.a3_in1k](https://huggingface.co/timm/resnext50_32x4d.a3_in1k)|224 |79.25|94.31|25.0 |4.3 |14.4 |2931 |
|[resnet50.fb_ssl_yfcc100m_ft_in1k](https://huggingface.co/timm/resnet50.fb_ssl_yfcc100m_ft_in1k)|224 |79.22|94.84|25.6 |4.1 |11.1 |3451 |
|[resnet33ts.ra2_in1k](https://huggingface.co/timm/resnet33ts.ra2_in1k)|256 |79.21|94.56|19.7 |4.8 |11.7 |3392 |
|[resnet50d.gluon_in1k](https://huggingface.co/timm/resnet50d.gluon_in1k)|224 |79.07|94.48|25.6 |4.4 |11.9 |3162 |
|[resnet50.ram_in1k](https://huggingface.co/timm/resnet50.ram_in1k)|224 |79.03|94.38|25.6 |4.1 |11.1 |3453 |
|[resnet50.am_in1k](https://huggingface.co/timm/resnet50.am_in1k)|224 |79.01|94.39|25.6 |4.1 |11.1 |3461 |
|[resnet32ts.ra2_in1k](https://huggingface.co/timm/resnet32ts.ra2_in1k)|256 |79.01|94.37|18.0 |4.6 |11.6 |3440 |
|[ecaresnet26t.ra2_in1k](https://huggingface.co/timm/ecaresnet26t.ra2_in1k)|256 |78.9 |94.54|16.0 |3.4 |10.5 |3421 |
|[resnet152.a3_in1k](https://huggingface.co/timm/resnet152.a3_in1k)|160 |78.89|94.11|60.2 |5.9 |11.5 |2745 |
|[wide_resnet101_2.tv_in1k](https://huggingface.co/timm/wide_resnet101_2.tv_in1k)|224 |78.84|94.28|126.9 |22.8 |21.2 |1079 |
|[seresnext26d_32x4d.bt_in1k](https://huggingface.co/timm/seresnext26d_32x4d.bt_in1k)|288 |78.83|94.24|16.8 |4.5 |16.8 |2251 |
|[resnet50.ra_in1k](https://huggingface.co/timm/resnet50.ra_in1k)|224 |78.81|94.32|25.6 |4.1 |11.1 |3454 |
|[seresnext26t_32x4d.bt_in1k](https://huggingface.co/timm/seresnext26t_32x4d.bt_in1k)|288 |78.74|94.33|16.8 |4.5 |16.7 |2264 |
|[resnet50s.gluon_in1k](https://huggingface.co/timm/resnet50s.gluon_in1k)|224 |78.72|94.23|25.7 |5.5 |13.5 |2796 |
|[resnet50d.a3_in1k](https://huggingface.co/timm/resnet50d.a3_in1k)|224 |78.71|94.24|25.6 |4.4 |11.9 |3154 |
|[wide_resnet50_2.tv_in1k](https://huggingface.co/timm/wide_resnet50_2.tv_in1k)|224 |78.47|94.09|68.9 |11.4 |14.4 |1934 |
|[resnet50.bt_in1k](https://huggingface.co/timm/resnet50.bt_in1k)|224 |78.46|94.27|25.6 |4.1 |11.1 |3454 |
|[resnet34d.ra2_in1k](https://huggingface.co/timm/resnet34d.ra2_in1k)|288 |78.43|94.35|21.8 |6.5 |7.5 |3291 |
|[gcresnext26ts.ch_in1k](https://huggingface.co/timm/gcresnext26ts.ch_in1k)|288 |78.42|94.04|10.5 |3.1 |13.3 |3226 |
|[resnet26t.ra2_in1k](https://huggingface.co/timm/resnet26t.ra2_in1k)|320 |78.33|94.13|16.0 |5.2 |16.4 |2391 |
|[resnet152.tv_in1k](https://huggingface.co/timm/resnet152.tv_in1k)|224 |78.32|94.04|60.2 |11.6 |22.6 |1487 |
|[seresnext26ts.ch_in1k](https://huggingface.co/timm/seresnext26ts.ch_in1k)|288 |78.28|94.1 |10.4 |3.1 |13.3 |3062 |
|[bat_resnext26ts.ch_in1k](https://huggingface.co/timm/bat_resnext26ts.ch_in1k)|256 |78.25|94.1 |10.7 |2.5 |12.5 |3393 |
|[resnet50.a3_in1k](https://huggingface.co/timm/resnet50.a3_in1k)|224 |78.06|93.78|25.6 |4.1 |11.1 |3450 |
|[resnet50c.gluon_in1k](https://huggingface.co/timm/resnet50c.gluon_in1k)|224 |78.0 |93.99|25.6 |4.4 |11.9 |3286 |
|[eca_resnext26ts.ch_in1k](https://huggingface.co/timm/eca_resnext26ts.ch_in1k)|288 |78.0 |93.91|10.3 |3.1 |13.3 |3297 |
|[seresnext26t_32x4d.bt_in1k](https://huggingface.co/timm/seresnext26t_32x4d.bt_in1k)|224 |77.98|93.75|16.8 |2.7 |10.1 |3841 |
|[resnet34.a1_in1k](https://huggingface.co/timm/resnet34.a1_in1k)|288 |77.92|93.77|21.8 |6.1 |6.2 |3609 |
|[resnet101.a3_in1k](https://huggingface.co/timm/resnet101.a3_in1k)|160 |77.88|93.71|44.6 |4.0 |8.3 |3926 |
|[resnet26t.ra2_in1k](https://huggingface.co/timm/resnet26t.ra2_in1k)|256 |77.87|93.84|16.0 |3.4 |10.5 |3772 |
|[seresnext26ts.ch_in1k](https://huggingface.co/timm/seresnext26ts.ch_in1k)|256 |77.86|93.79|10.4 |2.4 |10.5 |4263 |
|[resnetrs50.tf_in1k](https://huggingface.co/timm/resnetrs50.tf_in1k)|160 |77.82|93.81|35.7 |2.3 |6.2 |5238 |
|[gcresnext26ts.ch_in1k](https://huggingface.co/timm/gcresnext26ts.ch_in1k)|256 |77.81|93.82|10.5 |2.4 |10.5 |4183 |
|[ecaresnet50t.a3_in1k](https://huggingface.co/timm/ecaresnet50t.a3_in1k)|160 |77.79|93.6 |25.6 |2.2 |6.0 |5329 |
|[resnext50_32x4d.a3_in1k](https://huggingface.co/timm/resnext50_32x4d.a3_in1k)|160 |77.73|93.32|25.0 |2.2 |7.4 |5576 |
|[resnext50_32x4d.tv_in1k](https://huggingface.co/timm/resnext50_32x4d.tv_in1k)|224 |77.61|93.7 |25.0 |4.3 |14.4 |2944 |
|[seresnext26d_32x4d.bt_in1k](https://huggingface.co/timm/seresnext26d_32x4d.bt_in1k)|224 |77.59|93.61|16.8 |2.7 |10.2 |3807 |
|[resnet50.gluon_in1k](https://huggingface.co/timm/resnet50.gluon_in1k)|224 |77.58|93.72|25.6 |4.1 |11.1 |3455 |
|[eca_resnext26ts.ch_in1k](https://huggingface.co/timm/eca_resnext26ts.ch_in1k)|256 |77.44|93.56|10.3 |2.4 |10.5 |4284 |
|[resnet26d.bt_in1k](https://huggingface.co/timm/resnet26d.bt_in1k)|288 |77.41|93.63|16.0 |4.3 |13.5 |2907 |
|[resnet101.tv_in1k](https://huggingface.co/timm/resnet101.tv_in1k)|224 |77.38|93.54|44.6 |7.8 |16.2 |2125 |
|[resnet50d.a3_in1k](https://huggingface.co/timm/resnet50d.a3_in1k)|160 |77.22|93.27|25.6 |2.2 |6.1 |5982 |
|[resnext26ts.ra2_in1k](https://huggingface.co/timm/resnext26ts.ra2_in1k)|288 |77.17|93.47|10.3 |3.1 |13.3 |3392 |
|[resnet34.a2_in1k](https://huggingface.co/timm/resnet34.a2_in1k)|288 |77.15|93.27|21.8 |6.1 |6.2 |3615 |
|[resnet34d.ra2_in1k](https://huggingface.co/timm/resnet34d.ra2_in1k)|224 |77.1 |93.37|21.8 |3.9 |4.5 |5436 |
|[seresnet50.a3_in1k](https://huggingface.co/timm/seresnet50.a3_in1k)|224 |77.02|93.07|28.1 |4.1 |11.1 |2952 |
|[resnext26ts.ra2_in1k](https://huggingface.co/timm/resnext26ts.ra2_in1k)|256 |76.78|93.13|10.3 |2.4 |10.5 |4410 |
|[resnet26d.bt_in1k](https://huggingface.co/timm/resnet26d.bt_in1k)|224 |76.7 |93.17|16.0 |2.6 |8.2 |4859 |
|[resnet34.bt_in1k](https://huggingface.co/timm/resnet34.bt_in1k)|288 |76.5 |93.35|21.8 |6.1 |6.2 |3617 |
|[resnet34.a1_in1k](https://huggingface.co/timm/resnet34.a1_in1k)|224 |76.42|92.87|21.8 |3.7 |3.7 |5984 |
|[resnet26.bt_in1k](https://huggingface.co/timm/resnet26.bt_in1k)|288 |76.35|93.18|16.0 |3.9 |12.2 |3331 |
|[resnet50.tv_in1k](https://huggingface.co/timm/resnet50.tv_in1k)|224 |76.13|92.86|25.6 |4.1 |11.1 |3457 |
|[resnet50.a3_in1k](https://huggingface.co/timm/resnet50.a3_in1k)|160 |75.96|92.5 |25.6 |2.1 |5.7 |6490 |
|[resnet34.a2_in1k](https://huggingface.co/timm/resnet34.a2_in1k)|224 |75.52|92.44|21.8 |3.7 |3.7 |5991 |
|[resnet26.bt_in1k](https://huggingface.co/timm/resnet26.bt_in1k)|224 |75.3 |92.58|16.0 |2.4 |7.4 |5583 |
|[resnet34.bt_in1k](https://huggingface.co/timm/resnet34.bt_in1k)|224 |75.16|92.18|21.8 |3.7 |3.7 |5994 |
|[seresnet50.a3_in1k](https://huggingface.co/timm/seresnet50.a3_in1k)|160 |75.1 |92.08|28.1 |2.1 |5.7 |5513 |
|[resnet34.gluon_in1k](https://huggingface.co/timm/resnet34.gluon_in1k)|224 |74.57|91.98|21.8 |3.7 |3.7 |5984 |
|[resnet18d.ra2_in1k](https://huggingface.co/timm/resnet18d.ra2_in1k)|288 |73.81|91.83|11.7 |3.4 |5.4 |5196 |
|[resnet34.tv_in1k](https://huggingface.co/timm/resnet34.tv_in1k)|224 |73.32|91.42|21.8 |3.7 |3.7 |5979 |
|[resnet18.fb_swsl_ig1b_ft_in1k](https://huggingface.co/timm/resnet18.fb_swsl_ig1b_ft_in1k)|224 |73.28|91.73|11.7 |1.8 |2.5 |10213 |
|[resnet18.a1_in1k](https://huggingface.co/timm/resnet18.a1_in1k)|288 |73.16|91.03|11.7 |3.0 |4.1 |6050 |
|[resnet34.a3_in1k](https://huggingface.co/timm/resnet34.a3_in1k)|224 |72.98|91.11|21.8 |3.7 |3.7 |5967 |
|[resnet18.fb_ssl_yfcc100m_ft_in1k](https://huggingface.co/timm/resnet18.fb_ssl_yfcc100m_ft_in1k)|224 |72.6 |91.42|11.7 |1.8 |2.5 |10213 |
|[resnet18.a2_in1k](https://huggingface.co/timm/resnet18.a2_in1k)|288 |72.37|90.59|11.7 |3.0 |4.1 |6051 |
|[resnet14t.c3_in1k](https://huggingface.co/timm/resnet14t.c3_in1k)|224 |72.26|90.31|10.1 |1.7 |5.8 |7026 |
|[resnet18d.ra2_in1k](https://huggingface.co/timm/resnet18d.ra2_in1k)|224 |72.26|90.68|11.7 |2.1 |3.3 |8707 |
|[resnet18.a1_in1k](https://huggingface.co/timm/resnet18.a1_in1k)|224 |71.49|90.07|11.7 |1.8 |2.5 |10187 |
|[resnet14t.c3_in1k](https://huggingface.co/timm/resnet14t.c3_in1k)|176 |71.31|89.69|10.1 |1.1 |3.6 |10970 |
|[resnet18.gluon_in1k](https://huggingface.co/timm/resnet18.gluon_in1k)|224 |70.84|89.76|11.7 |1.8 |2.5 |10210 |
|[resnet18.a2_in1k](https://huggingface.co/timm/resnet18.a2_in1k)|224 |70.64|89.47|11.7 |1.8 |2.5 |10194 |
|[resnet34.a3_in1k](https://huggingface.co/timm/resnet34.a3_in1k)|160 |70.56|89.52|21.8 |1.9 |1.9 |10737 |
|[resnet18.tv_in1k](https://huggingface.co/timm/resnet18.tv_in1k)|224 |69.76|89.07|11.7 |1.8 |2.5 |10205 |
|[resnet10t.c3_in1k](https://huggingface.co/timm/resnet10t.c3_in1k)|224 |68.34|88.03|5.4 |1.1 |2.4 |13079 |
|[resnet18.a3_in1k](https://huggingface.co/timm/resnet18.a3_in1k)|224 |68.25|88.17|11.7 |1.8 |2.5 |10167 |
|[resnet10t.c3_in1k](https://huggingface.co/timm/resnet10t.c3_in1k)|176 |66.71|86.96|5.4 |0.7 |1.5 |20327 |
|[resnet18.a3_in1k](https://huggingface.co/timm/resnet18.a3_in1k)|160 |65.66|86.26|11.7 |0.9 |1.3 |18229 |
## Citation
```bibtex
@article{He2015,
author = {Kaiming He and Xiangyu Zhang and Shaoqing Ren and Jian Sun},
title = {Deep Residual Learning for Image Recognition},
journal = {arXiv preprint arXiv:1512.03385},
year = {2015}
}
```
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}
```
| 37,871 | [
[
-0.06683349609375,
-0.0164337158203125,
0.001712799072265625,
0.0307159423828125,
-0.0333251953125,
-0.00803375244140625,
-0.00930023193359375,
-0.0288848876953125,
0.087646484375,
0.0194244384765625,
-0.047576904296875,
-0.038848876953125,
-0.04595947265625,
-0.0015211105346679688,
0.026031494140625,
0.06396484375,
0.0005717277526855469,
-0.005390167236328125,
0.0180511474609375,
-0.0164031982421875,
-0.0028820037841796875,
-0.022918701171875,
-0.07525634765625,
-0.01221466064453125,
0.030853271484375,
0.0131988525390625,
0.050994873046875,
0.046905517578125,
0.027923583984375,
0.04498291015625,
-0.017669677734375,
0.021514892578125,
-0.005092620849609375,
-0.01122283935546875,
0.04791259765625,
-0.0288238525390625,
-0.06976318359375,
-0.0018396377563476562,
0.053375244140625,
0.047119140625,
0.006114959716796875,
0.02679443359375,
0.0296783447265625,
0.04461669921875,
0.003002166748046875,
-0.0030269622802734375,
0.0014696121215820312,
0.0090484619140625,
-0.0208740234375,
0.006061553955078125,
-0.0038585662841796875,
-0.055023193359375,
0.01387786865234375,
-0.0474853515625,
-0.00463104248046875,
0.00010019540786743164,
0.100830078125,
-0.009490966796875,
-0.0176239013671875,
0.006439208984375,
0.01136016845703125,
0.055419921875,
-0.06146240234375,
0.02557373046875,
0.044342041015625,
-0.0014219284057617188,
-0.01422119140625,
-0.0462646484375,
-0.037261962890625,
0.01027679443359375,
-0.032470703125,
0.0250091552734375,
-0.0224151611328125,
-0.016448974609375,
0.0278167724609375,
0.0243072509765625,
-0.035552978515625,
-0.01053619384765625,
-0.0276031494140625,
-0.006298065185546875,
0.05377197265625,
0.005283355712890625,
0.05157470703125,
-0.0274658203125,
-0.03741455078125,
-0.006801605224609375,
-0.01190185546875,
0.034698486328125,
0.019439697265625,
0.00870513916015625,
-0.0833740234375,
0.0325927734375,
0.009063720703125,
0.0168609619140625,
0.0276336669921875,
-0.00917816162109375,
0.06231689453125,
-0.007259368896484375,
-0.040191650390625,
-0.036712646484375,
0.0799560546875,
0.050018310546875,
0.021270751953125,
-0.0077362060546875,
-0.0009045600891113281,
-0.0108489990234375,
-0.03131103515625,
-0.0694580078125,
-0.0038623809814453125,
0.02069091796875,
-0.04168701171875,
-0.01654052734375,
0.0246429443359375,
-0.068359375,
-0.0040740966796875,
-0.00701904296875,
0.00522613525390625,
-0.05645751953125,
-0.033905029296875,
0.0008993148803710938,
-0.0168304443359375,
0.03997802734375,
0.0169219970703125,
-0.023345947265625,
0.031646728515625,
0.0034160614013671875,
0.06402587890625,
0.0229949951171875,
-0.0037021636962890625,
-0.01512908935546875,
0.0025463104248046875,
-0.0274200439453125,
0.0264739990234375,
0.0130157470703125,
-0.013885498046875,
-0.025115966796875,
0.031005859375,
-0.020111083984375,
-0.01558685302734375,
0.047454833984375,
0.0227813720703125,
0.0125885009765625,
-0.0208587646484375,
-0.019195556640625,
-0.0157318115234375,
0.02581787109375,
-0.04278564453125,
0.0745849609375,
0.0289154052734375,
-0.08380126953125,
0.01287841796875,
-0.0362548828125,
0.0005216598510742188,
-0.0222015380859375,
0.00650787353515625,
-0.0655517578125,
0.003292083740234375,
0.0162506103515625,
0.052734375,
-0.0157470703125,
-0.01334381103515625,
-0.0260467529296875,
0.004817962646484375,
0.0309295654296875,
0.01091766357421875,
0.06622314453125,
0.024627685546875,
-0.0328369140625,
-0.016754150390625,
-0.0548095703125,
0.03265380859375,
0.031494140625,
-0.0005040168762207031,
-0.0034999847412109375,
-0.058441162109375,
0.002605438232421875,
0.04620361328125,
0.0205535888671875,
-0.05487060546875,
0.0198516845703125,
-0.01305389404296875,
0.0248870849609375,
0.047576904296875,
0.001773834228515625,
0.0123748779296875,
-0.052337646484375,
0.045623779296875,
-0.0026950836181640625,
0.0210113525390625,
0.0010251998901367188,
-0.03125,
-0.056060791015625,
-0.05694580078125,
0.016876220703125,
0.03173828125,
-0.0301055908203125,
0.06597900390625,
0.01026153564453125,
-0.0452880859375,
-0.047821044921875,
0.003841400146484375,
0.04193115234375,
0.0153656005859375,
0.00643157958984375,
-0.02447509765625,
-0.055572509765625,
-0.07061767578125,
-0.02532958984375,
0.00930023193359375,
-0.0030384063720703125,
0.051788330078125,
0.031494140625,
-0.01383209228515625,
0.039764404296875,
-0.0285491943359375,
-0.0184326171875,
-0.0098419189453125,
-0.00814056396484375,
0.034088134765625,
0.059173583984375,
0.07562255859375,
-0.05487060546875,
-0.072509765625,
0.01165008544921875,
-0.08477783203125,
-0.0050811767578125,
-0.0005092620849609375,
-0.018829345703125,
0.031219482421875,
0.0186004638671875,
-0.06683349609375,
0.060028076171875,
0.0297393798828125,
-0.06500244140625,
0.034088134765625,
-0.027740478515625,
0.042877197265625,
-0.079345703125,
0.01953125,
0.020965576171875,
-0.01910400390625,
-0.043426513671875,
0.0026264190673828125,
-0.00800323486328125,
0.01139068603515625,
-0.04193115234375,
0.059326171875,
-0.05328369140625,
-0.0010919570922851562,
0.01226806640625,
0.006641387939453125,
-0.002895355224609375,
0.03302001953125,
-0.004207611083984375,
0.04486083984375,
0.06707763671875,
-0.01192474365234375,
0.0248565673828125,
0.032562255859375,
0.0025653839111328125,
0.05755615234375,
-0.04583740234375,
0.00620269775390625,
0.0019016265869140625,
0.03350830078125,
-0.07537841796875,
-0.0308990478515625,
0.042327880859375,
-0.06365966796875,
0.048095703125,
-0.0205230712890625,
-0.0199127197265625,
-0.06231689453125,
-0.06561279296875,
0.01934814453125,
0.04791259765625,
-0.04302978515625,
0.026885986328125,
0.01552581787109375,
-0.00411224365234375,
-0.036865234375,
-0.05145263671875,
0.0062408447265625,
-0.031829833984375,
-0.062103271484375,
0.033477783203125,
0.0254364013671875,
-0.01387786865234375,
0.0085906982421875,
-0.011199951171875,
-0.0117340087890625,
-0.0169219970703125,
0.0440673828125,
0.026336669921875,
-0.022979736328125,
-0.0309295654296875,
-0.0298309326171875,
-0.021392822265625,
-0.0051727294921875,
-0.0078277587890625,
0.03741455078125,
-0.0335693359375,
0.007205963134765625,
-0.11004638671875,
0.01013946533203125,
0.06689453125,
-0.001888275146484375,
0.0731201171875,
0.05670166015625,
-0.03509521484375,
0.01360321044921875,
-0.033538818359375,
-0.01568603515625,
-0.039031982421875,
-0.017974853515625,
-0.053314208984375,
-0.04400634765625,
0.06842041015625,
0.006031036376953125,
-0.0098876953125,
0.05804443359375,
0.011474609375,
-0.019256591796875,
0.06317138671875,
0.034576416015625,
-0.0036945343017578125,
0.042510986328125,
-0.0626220703125,
0.008636474609375,
-0.061431884765625,
-0.054931640625,
-0.016815185546875,
-0.040985107421875,
-0.04376220703125,
-0.0265960693359375,
0.017120361328125,
0.0302886962890625,
-0.018280029296875,
0.04461669921875,
-0.041473388671875,
0.0027599334716796875,
0.0227813720703125,
0.040435791015625,
-0.01467132568359375,
-0.0106964111328125,
-0.0081939697265625,
-0.0262451171875,
-0.03912353515625,
-0.02813720703125,
0.058837890625,
0.047332763671875,
0.03228759765625,
0.00830841064453125,
0.041229248046875,
0.00611114501953125,
0.01354217529296875,
-0.0218963623046875,
0.05078125,
0.00414276123046875,
-0.034332275390625,
-0.026641845703125,
-0.0297698974609375,
-0.08148193359375,
0.0137786865234375,
-0.034454345703125,
-0.06561279296875,
-0.01334381103515625,
-0.0041961669921875,
-0.027557373046875,
0.055023193359375,
-0.04595947265625,
0.0469970703125,
-0.004192352294921875,
-0.04144287109375,
-0.004756927490234375,
-0.05914306640625,
0.004730224609375,
0.030303955078125,
0.004001617431640625,
-0.0011539459228515625,
-0.00310516357421875,
0.0567626953125,
-0.06121826171875,
0.04339599609375,
-0.02520751953125,
0.0096588134765625,
0.030059814453125,
-0.0015020370483398438,
0.030120849609375,
0.0009255409240722656,
-0.01329803466796875,
-0.00893402099609375,
0.0092620849609375,
-0.06280517578125,
-0.0232391357421875,
0.04925537109375,
-0.055572509765625,
-0.0292510986328125,
-0.05059814453125,
-0.019805908203125,
0.00659942626953125,
0.0013885498046875,
0.03656005859375,
0.04852294921875,
-0.003002166748046875,
0.017364501953125,
0.04095458984375,
-0.03167724609375,
0.038909912109375,
-0.01165008544921875,
0.0007038116455078125,
-0.042144775390625,
0.054412841796875,
0.003307342529296875,
-0.0007348060607910156,
-0.0017347335815429688,
-0.00011211633682250977,
-0.031646728515625,
-0.016571044921875,
-0.0232391357421875,
0.0570068359375,
-0.0112152099609375,
-0.021728515625,
-0.04583740234375,
-0.0257415771484375,
-0.04229736328125,
-0.0330810546875,
-0.03289794921875,
-0.026397705078125,
-0.024688720703125,
0.00119781494140625,
0.05328369140625,
0.0653076171875,
-0.0280303955078125,
0.0297698974609375,
-0.039031982421875,
0.02349853515625,
0.006542205810546875,
0.043365478515625,
-0.02679443359375,
-0.049560546875,
0.0030612945556640625,
-0.0026760101318359375,
-0.005558013916015625,
-0.0626220703125,
0.050811767578125,
-0.0005021095275878906,
0.0282745361328125,
0.03131103515625,
-0.0163116455078125,
0.05352783203125,
-0.0010271072387695312,
0.0350341796875,
0.04595947265625,
-0.05511474609375,
0.0236358642578125,
-0.03515625,
0.0003807544708251953,
0.0214385986328125,
0.0146026611328125,
-0.028228759765625,
-0.0260467529296875,
-0.06536865234375,
-0.0294036865234375,
0.0548095703125,
0.0089569091796875,
-0.0029010772705078125,
-0.0015306472778320312,
0.0516357421875,
-0.006809234619140625,
0.004673004150390625,
-0.040008544921875,
-0.06939697265625,
-0.0090789794921875,
-0.01200103759765625,
0.0045013427734375,
-0.0027523040771484375,
0.0037441253662109375,
-0.0499267578125,
0.04937744140625,
0.005161285400390625,
0.03875732421875,
0.01361846923828125,
0.00290679931640625,
0.004161834716796875,
-0.022125244140625,
0.04620361328125,
0.028228759765625,
-0.01445770263671875,
-0.0115509033203125,
0.0269622802734375,
-0.037811279296875,
0.007335662841796875,
0.01715087890625,
0.0014476776123046875,
0.00666046142578125,
0.004726409912109375,
0.03717041015625,
0.0258331298828125,
-0.0052032470703125,
0.039337158203125,
-0.0194549560546875,
-0.042510986328125,
-0.0155792236328125,
-0.0170135498046875,
0.020721435546875,
0.032440185546875,
0.0250091552734375,
0.003162384033203125,
-0.0308074951171875,
-0.028045654296875,
0.04046630859375,
0.055511474609375,
-0.03082275390625,
-0.029022216796875,
0.044830322265625,
-0.0004706382751464844,
-0.01488494873046875,
0.0286102294921875,
-0.0084991455078125,
-0.051513671875,
0.07635498046875,
0.023468017578125,
0.045318603515625,
-0.0380859375,
0.0077056884765625,
0.064697265625,
-0.0016469955444335938,
0.0164337158203125,
0.02618408203125,
0.035552978515625,
-0.02362060546875,
-0.00670623779296875,
-0.040740966796875,
0.014434814453125,
0.037200927734375,
-0.0293731689453125,
0.0223541259765625,
-0.053863525390625,
-0.0252532958984375,
0.00629425048828125,
0.03631591796875,
-0.046600341796875,
0.026397705078125,
-0.0004646778106689453,
0.0804443359375,
-0.062164306640625,
0.065185546875,
0.06561279296875,
-0.04168701171875,
-0.06365966796875,
-0.0011281967163085938,
0.0081634521484375,
-0.0626220703125,
0.0325927734375,
0.005603790283203125,
0.0016269683837890625,
-0.0009775161743164062,
-0.0367431640625,
-0.0516357421875,
0.10150146484375,
0.028900146484375,
-0.001094818115234375,
0.0194091796875,
-0.0335693359375,
0.0276031494140625,
-0.01242828369140625,
0.04376220703125,
0.0282745361328125,
0.038177490234375,
0.01129913330078125,
-0.06732177734375,
0.025909423828125,
-0.032012939453125,
-0.01027679443359375,
0.0230712890625,
-0.098388671875,
0.06842041015625,
-0.0161285400390625,
-0.00217437744140625,
0.0192413330078125,
0.049285888671875,
0.02410888671875,
-0.0033931732177734375,
0.0166778564453125,
0.0667724609375,
0.035064697265625,
-0.0191497802734375,
0.07696533203125,
-0.0161895751953125,
0.042144775390625,
0.01520538330078125,
0.04229736328125,
0.0258941650390625,
0.030364990234375,
-0.044158935546875,
0.0189971923828125,
0.060577392578125,
-0.0038051605224609375,
0.00962066650390625,
0.0222015380859375,
-0.031494140625,
-0.015838623046875,
-0.01678466796875,
-0.051788330078125,
0.0172271728515625,
0.00955963134765625,
-0.00992584228515625,
-0.01050567626953125,
-0.0027713775634765625,
0.0177001953125,
0.0226593017578125,
-0.019805908203125,
0.037384033203125,
0.005126953125,
-0.030059814453125,
0.033477783203125,
-0.0029048919677734375,
0.08062744140625,
-0.0275115966796875,
0.01189422607421875,
-0.024566650390625,
0.02313232421875,
-0.01837158203125,
-0.08258056640625,
0.0243682861328125,
-0.0062408447265625,
0.00616455078125,
-0.0171356201171875,
0.047882080078125,
-0.0262451171875,
-0.0248870849609375,
0.0290374755859375,
0.0292205810546875,
0.03790283203125,
0.021575927734375,
-0.08197021484375,
0.0194854736328125,
0.007717132568359375,
-0.0458984375,
0.032928466796875,
0.0367431640625,
0.029571533203125,
0.057586669921875,
0.023284912109375,
0.0246734619140625,
0.01751708984375,
-0.0302886962890625,
0.055145263671875,
-0.047393798828125,
-0.034515380859375,
-0.06011962890625,
0.040802001953125,
-0.0310516357421875,
-0.040557861328125,
0.05572509765625,
0.0413818359375,
0.0265960693359375,
0.0005774497985839844,
0.05206298828125,
-0.042449951171875,
0.0379638671875,
-0.0196533203125,
0.0577392578125,
-0.0489501953125,
-0.020538330078125,
-0.01641845703125,
-0.043914794921875,
-0.031280517578125,
0.06536865234375,
-0.00876617431640625,
0.0162506103515625,
0.0221099853515625,
0.049957275390625,
0.00644683837890625,
-0.00844573974609375,
0.0011949539184570312,
0.0134124755859375,
-0.0113525390625,
0.0675048828125,
0.03887939453125,
-0.0570068359375,
0.00501251220703125,
-0.034454345703125,
-0.0215911865234375,
-0.0288238525390625,
-0.055206298828125,
-0.087646484375,
-0.04998779296875,
-0.0400390625,
-0.05078125,
-0.02069091796875,
0.09100341796875,
0.05987548828125,
-0.04376220703125,
-0.01210784912109375,
0.0104522705078125,
0.007305145263671875,
-0.01108551025390625,
-0.0160675048828125,
0.04010009765625,
0.01043701171875,
-0.074951171875,
-0.032196044921875,
0.0092926025390625,
0.0455322265625,
0.0297393798828125,
-0.03619384765625,
-0.019744873046875,
-0.0042724609375,
0.026214599609375,
0.0653076171875,
-0.061492919921875,
-0.022430419921875,
0.002071380615234375,
-0.037872314453125,
0.009918212890625,
0.019866943359375,
-0.0341796875,
-0.00848388671875,
0.037811279296875,
0.028076171875,
0.05413818359375,
0.006481170654296875,
0.0124359130859375,
-0.029998779296875,
0.0408935546875,
-0.000972747802734375,
0.0252227783203125,
0.0153961181640625,
-0.0214080810546875,
0.05712890625,
0.04095458984375,
-0.030853271484375,
-0.07562255859375,
-0.01318359375,
-0.0992431640625,
-0.00464630126953125,
0.048919677734375,
-0.0046844482421875,
-0.030120849609375,
0.0292510986328125,
-0.03607177734375,
0.03936767578125,
-0.0177764892578125,
0.0185699462890625,
0.018402099609375,
-0.027557373046875,
-0.025054931640625,
-0.043121337890625,
0.046905517578125,
0.0264739990234375,
-0.050872802734375,
-0.02789306640625,
0.0008292198181152344,
0.0224456787109375,
0.012908935546875,
0.055938720703125,
-0.02850341796875,
0.01132965087890625,
-0.0070648193359375,
0.0170440673828125,
0.0006313323974609375,
0.01226806640625,
-0.023590087890625,
-0.01074981689453125,
-0.0184173583984375,
-0.0491943359375
]
] |
KoboldAI/GPT-NeoX-20B-Skein | 2022-09-26T19:19:21.000Z | [
"transformers",
"pytorch",
"gpt_neox",
"text-generation",
"en",
"arxiv:2204.06745",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | KoboldAI | null | null | KoboldAI/GPT-NeoX-20B-Skein | 9 | 7,367 | transformers | 2022-08-18T23:13:05 | ---
language: en
license: apache-2.0
---
# GPT-NeoX-20B-Skein
## Model description
Skein is a series of hybrid story generation models intended for use in both text adventure writing and normal novel-style writing. The models are known to possess a strong second person bias. For inquiries, please contact the KoboldAI community.
The name comes from the Integrated Development Environment for the Inform 7 programming language, which calls a dialogue tree a "skein". Inform 6 and 7 were used to create some of the interactive fiction in the dataset.
## Training procedure
GPT-NeoX-20B-Skein was trained on a TPUv3-32 TPU pod using a heavily modified version of Ben Wang's Mesh Transformer JAX library, the original version of which was used by EleutherAI to train their GPT-J-6B model. The training hyperparameters and statistics can be found [here](https://wandb.ai/ve-forbryderne/skein-20b?workspace=user-ve-forbryderne).
## Training data
The data are mostly comprised of light novels from the dataset of the [KoboldAI/GPT-Neo-2.7B-Horni-LN](https://huggingface.co/KoboldAI/GPT-Neo-2.7B-Horni-LN) model and assorted interactive fiction. The dataset uses `[Themes: <comma-separated list of genres>]` for tagging. For more details, consult [this document](https://wandb.ai/ve-forbryderne/skein/runs/files/files/datasets/README.txt).
## Limitations and biases
Based on known problems with NLP technology, potential relevant factors include bias (gender, profession, race and religion).
## Citation details
The GPT-NeoX-20B model weights:
```bibtex
@inproceedings{gpt-neox-20b,
title={{GPT-NeoX-20B}: An Open-Source Autoregressive Language Model},
author={Black, Sid and Biderman, Stella and Hallahan, Eric and Anthony, Quentin and Gao, Leo and Golding, Laurence and He, Horace and Leahy, Connor and McDonell, Kyle and Phang, Jason and Pieler, Michael and Prashanth, USVSN Sai and Purohit, Shivanshu and Reynolds, Laria and Tow, Jonathan and Wang, Ben and Weinbach, Samuel},
booktitle={Proceedings of the ACL Workshop on Challenges \& Perspectives in Creating Large Language Models},
url={https://arxiv.org/abs/2204.06745},
year={2022}
}
```
The Mesh Transformer JAX library:
```bibtex
@misc{mesh-transformer-jax,
author = {Wang, Ben},
title = {{Mesh-Transformer-JAX: Model-Parallel Implementation of Transformer Language Model with JAX}},
howpublished = {\url{https://github.com/kingoflolz/mesh-transformer-jax}},
year = 2021,
month = May
}
```
| 2,470 | [
[
-0.034515380859375,
-0.0462646484375,
0.03448486328125,
0.0101470947265625,
-0.01531982421875,
0.00275421142578125,
-0.016021728515625,
-0.05035400390625,
0.0017919540405273438,
0.04754638671875,
-0.04345703125,
-0.033294677734375,
-0.0273895263671875,
0.00623321533203125,
-0.0262603759765625,
0.0947265625,
0.00543975830078125,
-0.01114654541015625,
-0.0034351348876953125,
-0.00997161865234375,
-0.0088653564453125,
-0.035400390625,
-0.0516357421875,
-0.0216827392578125,
0.03857421875,
-0.01468658447265625,
0.0582275390625,
0.059478759765625,
0.01044464111328125,
0.023590087890625,
-0.0186004638671875,
-0.0188140869140625,
-0.0276031494140625,
-0.0102691650390625,
-0.0184478759765625,
-0.03399658203125,
-0.034149169921875,
0.00852203369140625,
0.052459716796875,
0.041290283203125,
0.0016126632690429688,
0.0166015625,
-0.01538848876953125,
0.02288818359375,
-0.01416778564453125,
-0.002044677734375,
-0.0283050537109375,
-0.016387939453125,
-0.038818359375,
0.023284912109375,
-0.02618408203125,
-0.0188140869140625,
0.0205535888671875,
-0.04864501953125,
0.052734375,
0.020416259765625,
0.07666015625,
-0.00547027587890625,
-0.048858642578125,
-0.0239105224609375,
-0.06298828125,
0.06939697265625,
-0.0877685546875,
0.030426025390625,
0.0113067626953125,
0.01326751708984375,
-0.005889892578125,
-0.0572509765625,
-0.06317138671875,
-0.025146484375,
-0.0225982666015625,
0.014434814453125,
0.00408172607421875,
0.0089263916015625,
0.04443359375,
0.05059814453125,
-0.0665283203125,
-0.0101470947265625,
-0.0357666015625,
-0.013580322265625,
0.056854248046875,
0.018280029296875,
0.0248870849609375,
-0.024932861328125,
-0.0230255126953125,
-0.04412841796875,
-0.033782958984375,
-0.00849151611328125,
0.06317138671875,
0.007724761962890625,
-0.02264404296875,
0.041015625,
-0.00302886962890625,
0.046051025390625,
-0.005084991455078125,
-0.0208892822265625,
0.0305938720703125,
-0.040283203125,
0.0015573501586914062,
-0.01143646240234375,
0.07940673828125,
0.019317626953125,
0.01056671142578125,
-0.0113983154296875,
-0.0125885009765625,
0.01303863525390625,
0.02252197265625,
-0.050506591796875,
-0.027618408203125,
0.0127716064453125,
-0.03753662109375,
-0.0292816162109375,
0.0038166046142578125,
-0.064208984375,
-0.025115966796875,
-0.0246429443359375,
0.0223236083984375,
-0.0276947021484375,
-0.0506591796875,
0.0003287792205810547,
-0.0260467529296875,
0.016326904296875,
0.0007271766662597656,
-0.059356689453125,
0.0294647216796875,
0.031341552734375,
0.052978515625,
-0.01490020751953125,
-0.018768310546875,
0.01206207275390625,
0.01312255859375,
-0.01291656494140625,
0.039886474609375,
-0.0307159423828125,
-0.027496337890625,
-0.0122833251953125,
0.02337646484375,
-0.01132965087890625,
-0.0121612548828125,
0.040283203125,
-0.0208892822265625,
0.041900634765625,
0.00447845458984375,
-0.0313720703125,
-0.02276611328125,
-0.00780487060546875,
-0.057098388671875,
0.09954833984375,
0.03717041015625,
-0.06884765625,
0.0183258056640625,
-0.0352783203125,
-0.007709503173828125,
0.01213836669921875,
-0.004802703857421875,
-0.033111572265625,
0.01253509521484375,
-0.0006914138793945312,
-0.00021398067474365234,
-0.044219970703125,
0.0305938720703125,
-0.02947998046875,
-0.01273345947265625,
-0.0066375732421875,
-0.0279693603515625,
0.08184814453125,
0.0120086669921875,
-0.042816162109375,
-0.0005474090576171875,
-0.0726318359375,
0.003978729248046875,
0.0308990478515625,
-0.00684356689453125,
-0.0408935546875,
-0.0182952880859375,
0.0211639404296875,
0.027862548828125,
0.0100860595703125,
-0.0604248046875,
0.0286865234375,
-0.034881591796875,
0.0248260498046875,
0.049468994140625,
-0.01125335693359375,
0.036712646484375,
-0.0309906005859375,
0.047515869140625,
-0.01534271240234375,
0.0189971923828125,
0.00044417381286621094,
-0.04559326171875,
-0.0643310546875,
-0.000030219554901123047,
0.0206451416015625,
0.047882080078125,
-0.06494140625,
0.0152130126953125,
-0.0047760009765625,
-0.070556640625,
-0.045318603515625,
0.00009554624557495117,
0.0374755859375,
0.016510009765625,
0.0281219482421875,
-0.0014657974243164062,
-0.03802490234375,
-0.062103271484375,
-0.0285491943359375,
-0.0288543701171875,
0.007305145263671875,
0.038055419921875,
0.031402587890625,
-0.0281829833984375,
0.064208984375,
-0.0268096923828125,
-0.00041484832763671875,
-0.033599853515625,
0.0296783447265625,
0.039520263671875,
0.04119873046875,
0.04766845703125,
-0.06317138671875,
-0.047637939453125,
0.004638671875,
-0.032440185546875,
-0.0302734375,
-0.022308349609375,
-0.0127105712890625,
0.027130126953125,
0.0242919921875,
-0.0477294921875,
0.037811279296875,
0.0594482421875,
-0.048980712890625,
0.06829833984375,
-0.005138397216796875,
-0.00920867919921875,
-0.0960693359375,
0.009735107421875,
0.00022780895233154297,
-0.00868988037109375,
-0.05279541015625,
-0.00618743896484375,
-0.004940032958984375,
0.0027923583984375,
-0.016082763671875,
0.0731201171875,
-0.034423828125,
0.0081024169921875,
-0.01479339599609375,
-0.008819580078125,
-0.003459930419921875,
0.03570556640625,
0.0141448974609375,
0.04132080078125,
0.025177001953125,
-0.03326416015625,
0.030364990234375,
0.0168914794921875,
-0.007061004638671875,
0.0301055908203125,
-0.0648193359375,
0.0186614990234375,
0.00664520263671875,
0.01824951171875,
-0.055999755859375,
-0.00995635986328125,
0.0250701904296875,
-0.018280029296875,
0.0183868408203125,
-0.0173187255859375,
-0.0390625,
-0.048736572265625,
-0.0035037994384765625,
0.04718017578125,
0.03570556640625,
-0.0308380126953125,
0.059814453125,
0.00738525390625,
-0.024261474609375,
-0.0286712646484375,
-0.034881591796875,
-0.01450347900390625,
-0.01873779296875,
-0.03558349609375,
0.02630615234375,
-0.0284881591796875,
-0.0035533905029296875,
0.008941650390625,
0.0010356903076171875,
0.00640106201171875,
-0.01898193359375,
0.0172882080078125,
0.0283355712890625,
-0.0184478759765625,
-0.007045745849609375,
0.0134429931640625,
-0.01922607421875,
0.0017375946044921875,
-0.027801513671875,
0.057373046875,
-0.01407623291015625,
-0.00726318359375,
-0.0226898193359375,
0.03424072265625,
0.038330078125,
0.0002799034118652344,
0.06060791015625,
0.07562255859375,
-0.00634765625,
-0.007419586181640625,
-0.049041748046875,
-0.00986480712890625,
-0.0350341796875,
0.025970458984375,
-0.0247650146484375,
-0.0771484375,
0.0389404296875,
0.004932403564453125,
0.0156707763671875,
0.0654296875,
0.04425048828125,
0.0103607177734375,
0.08966064453125,
0.065673828125,
-0.0008487701416015625,
0.039794921875,
-0.024658203125,
0.0170745849609375,
-0.072021484375,
-0.0050811767578125,
-0.044464111328125,
-0.01116943359375,
-0.0631103515625,
-0.01418304443359375,
0.00988006591796875,
0.01251220703125,
-0.02911376953125,
0.049713134765625,
-0.04119873046875,
0.0163116455078125,
0.05450439453125,
-0.01806640625,
0.025482177734375,
-0.002597808837890625,
-0.023223876953125,
0.0008959770202636719,
-0.05255126953125,
-0.030364990234375,
0.08392333984375,
0.0259246826171875,
0.05255126953125,
0.00638580322265625,
0.050750732421875,
0.0019550323486328125,
0.0183563232421875,
-0.0281829833984375,
0.03448486328125,
-0.006214141845703125,
-0.07354736328125,
-0.0222930908203125,
-0.0443115234375,
-0.05609130859375,
0.00873565673828125,
-0.01094818115234375,
-0.06744384765625,
0.0408935546875,
0.0051116943359375,
-0.006969451904296875,
0.030059814453125,
-0.047271728515625,
0.07391357421875,
-0.01361846923828125,
0.001251220703125,
0.0019741058349609375,
-0.044952392578125,
0.032623291015625,
-0.00025725364685058594,
0.00930023193359375,
0.00615692138671875,
0.00795745849609375,
0.055206298828125,
-0.0216064453125,
0.061004638671875,
-0.0030994415283203125,
-0.01462554931640625,
0.017852783203125,
0.0040130615234375,
0.038970947265625,
0.0192718505859375,
0.00836181640625,
0.031341552734375,
0.0255584716796875,
-0.0255584716796875,
-0.0272979736328125,
0.041961669921875,
-0.0712890625,
-0.037445068359375,
-0.055755615234375,
-0.061981201171875,
0.009124755859375,
0.04034423828125,
0.03607177734375,
0.044952392578125,
0.003170013427734375,
0.01715087890625,
0.037200927734375,
-0.030548095703125,
0.0286865234375,
0.03912353515625,
-0.033843994140625,
-0.04864501953125,
0.07440185546875,
0.00629425048828125,
0.0435791015625,
-0.0019474029541015625,
0.0220184326171875,
-0.01824951171875,
-0.03533935546875,
-0.03643798828125,
0.058746337890625,
-0.0276641845703125,
-0.0002161264419555664,
-0.06951904296875,
-0.03521728515625,
-0.041717529296875,
-0.00452423095703125,
-0.02569580078125,
-0.033294677734375,
0.0003597736358642578,
0.01134490966796875,
0.027130126953125,
0.0650634765625,
0.00872039794921875,
0.03900146484375,
-0.041412353515625,
0.0250701904296875,
0.01201629638671875,
0.0230560302734375,
-0.0025577545166015625,
-0.038818359375,
-0.0232391357421875,
0.0042266845703125,
-0.00829315185546875,
-0.05126953125,
0.035552978515625,
0.0054168701171875,
0.048736572265625,
0.01067352294921875,
0.01427459716796875,
0.03021240234375,
-0.027740478515625,
0.06719970703125,
0.0021228790283203125,
-0.056060791015625,
0.0263519287109375,
-0.04443359375,
0.037567138671875,
0.021087646484375,
0.03619384765625,
-0.061065673828125,
-0.0577392578125,
-0.07379150390625,
-0.0723876953125,
0.07080078125,
0.0173797607421875,
0.00893402099609375,
0.000002562999725341797,
0.023681640625,
0.0172882080078125,
0.013671875,
-0.07989501953125,
-0.0253753662109375,
-0.025604248046875,
-0.02337646484375,
-0.01971435546875,
-0.028472900390625,
-0.0128936767578125,
-0.0168609619140625,
0.07275390625,
-0.01317596435546875,
0.037689208984375,
0.0136871337890625,
-0.01166534423828125,
0.0017709732055664062,
0.005474090576171875,
0.0312347412109375,
0.043365478515625,
-0.0192718505859375,
-0.0133514404296875,
0.0016508102416992188,
-0.06719970703125,
-0.006542205810546875,
0.026336669921875,
-0.03289794921875,
0.00994110107421875,
0.0308380126953125,
0.07672119140625,
-0.00995635986328125,
-0.0277862548828125,
0.02703857421875,
-0.00707244873046875,
-0.0295562744140625,
-0.02947998046875,
0.01047515869140625,
0.003772735595703125,
0.030303955078125,
0.0193023681640625,
-0.01751708984375,
0.007770538330078125,
-0.0248565673828125,
-0.01113128662109375,
0.0153045654296875,
-0.023773193359375,
-0.030792236328125,
0.041656494140625,
-0.007045745849609375,
-0.0107421875,
0.04864501953125,
-0.0251922607421875,
-0.0305328369140625,
0.0292816162109375,
0.072265625,
0.07476806640625,
-0.020599365234375,
0.03350830078125,
0.0401611328125,
0.03924560546875,
0.00304412841796875,
0.0133819580078125,
0.00794219970703125,
-0.049102783203125,
-0.0222930908203125,
-0.07012939453125,
-0.00931549072265625,
0.0244903564453125,
-0.037628173828125,
0.016021728515625,
-0.0304718017578125,
-0.0109405517578125,
-0.01343536376953125,
0.0128173828125,
-0.043701171875,
0.0099029541015625,
0.005710601806640625,
0.06219482421875,
-0.0758056640625,
0.06011962890625,
0.056121826171875,
-0.023834228515625,
-0.058990478515625,
-0.0110321044921875,
0.011871337890625,
-0.01152801513671875,
0.022430419921875,
-0.005336761474609375,
0.02337646484375,
0.0231170654296875,
-0.037261962890625,
-0.079345703125,
0.093994140625,
0.00925445556640625,
-0.03662109375,
-0.027130126953125,
0.020416259765625,
0.0457763671875,
-0.02545166015625,
0.0389404296875,
0.03900146484375,
0.0352783203125,
-0.00595855712890625,
-0.0706787109375,
0.0059814453125,
-0.039642333984375,
0.03228759765625,
0.0190887451171875,
-0.055511474609375,
0.07537841796875,
0.023956298828125,
-0.01291656494140625,
0.0290069580078125,
0.041961669921875,
0.0208587646484375,
0.0159759521484375,
0.051605224609375,
0.058380126953125,
0.03814697265625,
-0.01641845703125,
0.091064453125,
-0.01227569580078125,
0.031585693359375,
0.076416015625,
0.0056610107421875,
0.027557373046875,
0.01300811767578125,
-0.02008056640625,
0.05474853515625,
0.033843994140625,
-0.0117340087890625,
0.0215911865234375,
-0.005542755126953125,
-0.0168304443359375,
-0.016632080078125,
-0.006427764892578125,
-0.037445068359375,
0.01306915283203125,
0.0262603759765625,
-0.034698486328125,
-0.007659912109375,
0.004589080810546875,
0.01250457763671875,
-0.01708984375,
-0.002231597900390625,
0.05908203125,
0.014312744140625,
-0.0264892578125,
0.033782958984375,
-0.0072479248046875,
0.03863525390625,
-0.0609130859375,
0.0005555152893066406,
-0.00498199462890625,
0.01508331298828125,
-0.00093841552734375,
-0.040283203125,
0.0127410888671875,
-0.01076507568359375,
-0.041259765625,
-0.00457000732421875,
0.0506591796875,
-0.04302978515625,
-0.060333251953125,
0.0252532958984375,
0.02154541015625,
0.012664794921875,
0.010772705078125,
-0.073974609375,
0.0074920654296875,
0.0016794204711914062,
-0.037445068359375,
0.023040771484375,
0.034576416015625,
-0.02911376953125,
0.03204345703125,
0.06988525390625,
0.01605224609375,
0.00946044921875,
0.0122833251953125,
0.06988525390625,
-0.0528564453125,
-0.038726806640625,
-0.032623291015625,
0.046051025390625,
-0.020599365234375,
-0.04376220703125,
0.06854248046875,
0.034820556640625,
0.07635498046875,
-0.004161834716796875,
0.06390380859375,
-0.0380859375,
0.056243896484375,
-0.0212554931640625,
0.061676025390625,
-0.0312347412109375,
-0.00551605224609375,
-0.043182373046875,
-0.1085205078125,
-0.0005545616149902344,
0.0330810546875,
-0.026763916015625,
0.0212554931640625,
0.06292724609375,
0.066162109375,
-0.007259368896484375,
0.005828857421875,
0.0061187744140625,
0.036224365234375,
0.033660888671875,
0.031280517578125,
0.043792724609375,
-0.044769287109375,
0.03997802734375,
-0.044677734375,
-0.0197601318359375,
-0.0038509368896484375,
-0.064697265625,
-0.08685302734375,
-0.051910400390625,
-0.020172119140625,
-0.0416259765625,
0.0016183853149414062,
0.045318603515625,
0.03790283203125,
-0.06524658203125,
-0.023834228515625,
-0.0142822265625,
-0.005878448486328125,
-0.0226898193359375,
-0.0212249755859375,
0.0259246826171875,
-0.01152801513671875,
-0.0587158203125,
0.0091400146484375,
-0.0008797645568847656,
0.006160736083984375,
-0.024566650390625,
-0.01910400390625,
-0.0307159423828125,
-0.0014638900756835938,
0.0192718505859375,
0.0218963623046875,
-0.049652099609375,
-0.0108795166015625,
0.004344940185546875,
-0.0102386474609375,
-0.030303955078125,
0.0546875,
-0.049407958984375,
0.034637451171875,
0.04083251953125,
0.025482177734375,
0.04547119140625,
-0.00934600830078125,
0.059295654296875,
-0.0302734375,
0.01605224609375,
0.01462554931640625,
0.01467132568359375,
0.01146697998046875,
-0.02508544921875,
0.041412353515625,
0.052703857421875,
-0.0528564453125,
-0.0577392578125,
0.016510009765625,
-0.07470703125,
-0.00225067138671875,
0.09857177734375,
-0.0036945343017578125,
-0.031280517578125,
-0.0196075439453125,
-0.036376953125,
0.02166748046875,
-0.01052093505859375,
0.02935791015625,
0.0604248046875,
0.0195770263671875,
-0.03656005859375,
-0.06182861328125,
0.048095703125,
0.0117950439453125,
-0.037628173828125,
0.00408172607421875,
0.0232086181640625,
0.012451171875,
0.0328369140625,
0.02508544921875,
-0.0262451171875,
0.016815185546875,
0.011962890625,
0.0258026123046875,
-0.0139617919921875,
-0.0195465087890625,
-0.01241302490234375,
-0.02178955078125,
-0.027496337890625,
0.04351806640625
]
] |
sentence-transformers/msmarco-MiniLM-L-12-v3 | 2022-06-16T00:16:13.000Z | [
"sentence-transformers",
"pytorch",
"tf",
"jax",
"bert",
"feature-extraction",
"sentence-similarity",
"transformers",
"arxiv:1908.10084",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | sentence-similarity | sentence-transformers | null | null | sentence-transformers/msmarco-MiniLM-L-12-v3 | 14 | 7,347 | sentence-transformers | 2022-03-02T23:29:05 | ---
pipeline_tag: sentence-similarity
license: apache-2.0
tags:
- sentence-transformers
- feature-extraction
- sentence-similarity
- transformers
---
# sentence-transformers/msmarco-MiniLM-L-12-v3
This is a [sentence-transformers](https://www.SBERT.net) model: It maps sentences & paragraphs to a 384 dimensional dense vector space and can be used for tasks like clustering or semantic search.
## Usage (Sentence-Transformers)
Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:
```
pip install -U sentence-transformers
```
Then you can use the model like this:
```python
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]
model = SentenceTransformer('sentence-transformers/msmarco-MiniLM-L-12-v3')
embeddings = model.encode(sentences)
print(embeddings)
```
## Usage (HuggingFace Transformers)
Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
```python
from transformers import AutoTokenizer, AutoModel
import torch
#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
token_embeddings = model_output[0] #First element of model_output contains all token embeddings
input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']
# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('sentence-transformers/msmarco-MiniLM-L-12-v3')
model = AutoModel.from_pretrained('sentence-transformers/msmarco-MiniLM-L-12-v3')
# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
# Compute token embeddings
with torch.no_grad():
model_output = model(**encoded_input)
# Perform pooling. In this case, max pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
```
## Evaluation Results
For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=sentence-transformers/msmarco-MiniLM-L-12-v3)
## Full Model Architecture
```
SentenceTransformer(
(0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)
```
## Citing & Authors
This model was trained by [sentence-transformers](https://www.sbert.net/).
If you find this model helpful, feel free to cite our publication [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://arxiv.org/abs/1908.10084):
```bibtex
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "http://arxiv.org/abs/1908.10084",
}
``` | 3,688 | [
[
-0.02362060546875,
-0.049346923828125,
0.018463134765625,
0.021148681640625,
-0.021636962890625,
-0.03350830078125,
-0.0240631103515625,
-0.005863189697265625,
0.01488494873046875,
0.0256500244140625,
-0.04827880859375,
-0.0284423828125,
-0.049102783203125,
0.01238250732421875,
-0.0281219482421875,
0.07147216796875,
-0.006793975830078125,
0.0006661415100097656,
-0.0253143310546875,
-0.019256591796875,
-0.0198822021484375,
-0.0272979736328125,
-0.03411865234375,
-0.0129241943359375,
0.0258026123046875,
0.018768310546875,
0.03729248046875,
0.0289306640625,
0.019500732421875,
0.033721923828125,
-0.004398345947265625,
0.022308349609375,
-0.0237884521484375,
-0.01056671142578125,
0.006816864013671875,
-0.0318603515625,
-0.0040740966796875,
0.01435089111328125,
0.042816162109375,
0.029052734375,
-0.0129852294921875,
0.01496124267578125,
0.012054443359375,
0.023834228515625,
-0.027862548828125,
0.03399658203125,
-0.0452880859375,
0.0138397216796875,
0.0026950836181640625,
0.0036792755126953125,
-0.044219970703125,
-0.0090789794921875,
0.01328277587890625,
-0.0242767333984375,
0.0204620361328125,
0.0261688232421875,
0.08563232421875,
0.030242919921875,
-0.017822265625,
-0.03521728515625,
-0.01873779296875,
0.06341552734375,
-0.06317138671875,
0.0119171142578125,
0.0218658447265625,
0.00539398193359375,
0.00032329559326171875,
-0.0755615234375,
-0.056121826171875,
-0.01270294189453125,
-0.044403076171875,
0.0109100341796875,
-0.029388427734375,
0.0020122528076171875,
0.0189971923828125,
0.016510009765625,
-0.05413818359375,
-0.00518798828125,
-0.037384033203125,
-0.01806640625,
0.039398193359375,
0.0131683349609375,
0.0235443115234375,
-0.045166015625,
-0.033538818359375,
-0.0209503173828125,
-0.0165557861328125,
0.0031681060791015625,
0.006595611572265625,
0.0211334228515625,
-0.027191162109375,
0.066162109375,
0.0012178421020507812,
0.040435791015625,
-0.00716400146484375,
0.00988006591796875,
0.041290283203125,
-0.03753662109375,
-0.0268707275390625,
-0.00722503662109375,
0.0841064453125,
0.019927978515625,
0.0169219970703125,
0.0089569091796875,
-0.0148162841796875,
0.006671905517578125,
0.0159454345703125,
-0.059722900390625,
-0.0298004150390625,
0.012603759765625,
-0.035491943359375,
-0.03045654296875,
0.006256103515625,
-0.052337646484375,
-0.00504302978515625,
0.006252288818359375,
0.052764892578125,
-0.040191650390625,
0.01922607421875,
0.0103759765625,
-0.01549530029296875,
0.009124755859375,
-0.0238800048828125,
-0.05389404296875,
0.016754150390625,
0.011016845703125,
0.0797119140625,
0.00569915771484375,
-0.036895751953125,
-0.0218505859375,
-0.0099029541015625,
0.005146026611328125,
0.06036376953125,
-0.0216522216796875,
-0.004833221435546875,
0.0108642578125,
0.0230255126953125,
-0.047760009765625,
-0.0253143310546875,
0.042724609375,
-0.0204010009765625,
0.05450439453125,
0.01238250732421875,
-0.05535888671875,
-0.0046539306640625,
0.00605010986328125,
-0.039276123046875,
0.0853271484375,
0.00919342041015625,
-0.07379150390625,
0.00009399652481079102,
-0.060089111328125,
-0.0172119140625,
-0.0160675048828125,
-0.00402069091796875,
-0.05224609375,
0.010345458984375,
0.03570556640625,
0.047119140625,
0.006069183349609375,
0.01256561279296875,
-0.01641845703125,
-0.0303802490234375,
0.02679443359375,
-0.01910400390625,
0.0872802734375,
0.01611328125,
-0.0234527587890625,
0.0225830078125,
-0.0301513671875,
-0.012451171875,
0.0289459228515625,
-0.00646209716796875,
-0.01715087890625,
-0.00821685791015625,
0.01358795166015625,
0.030609130859375,
0.0172119140625,
-0.04302978515625,
0.01861572265625,
-0.04205322265625,
0.059844970703125,
0.05255126953125,
0.00007468461990356445,
0.046356201171875,
-0.02947998046875,
0.0198822021484375,
0.018524169921875,
0.005939483642578125,
-0.019012451171875,
-0.0390625,
-0.08306884765625,
-0.02972412109375,
0.0261077880859375,
0.0374755859375,
-0.0703125,
0.068603515625,
-0.032470703125,
-0.03680419921875,
-0.06561279296875,
-0.0016536712646484375,
0.00787353515625,
0.031524658203125,
0.046051025390625,
0.0016584396362304688,
-0.053070068359375,
-0.0767822265625,
-0.009490966796875,
0.004329681396484375,
0.00496673583984375,
0.01861572265625,
0.05816650390625,
-0.03759765625,
0.074951171875,
-0.05047607421875,
-0.04693603515625,
-0.03387451171875,
0.017791748046875,
0.0209503173828125,
0.0390625,
0.03753662109375,
-0.04388427734375,
-0.0341796875,
-0.042724609375,
-0.052398681640625,
-0.0026187896728515625,
-0.01788330078125,
-0.014312744140625,
0.0185546875,
0.038238525390625,
-0.061492919921875,
0.0286407470703125,
0.0504150390625,
-0.03826904296875,
0.0300750732421875,
-0.0234527587890625,
-0.019622802734375,
-0.09326171875,
-0.0020885467529296875,
-0.00494384765625,
-0.02008056640625,
-0.033599853515625,
0.0032672882080078125,
0.01059722900390625,
-0.00853729248046875,
-0.036529541015625,
0.037628173828125,
-0.0333251953125,
0.0088653564453125,
-0.003955841064453125,
0.0379638671875,
-0.0005941390991210938,
0.05902099609375,
-0.01436614990234375,
0.055450439453125,
0.033111572265625,
-0.038116455078125,
0.0207061767578125,
0.043487548828125,
-0.038360595703125,
0.00720977783203125,
-0.058563232421875,
0.003582000732421875,
-0.004001617431640625,
0.0256195068359375,
-0.08770751953125,
0.004138946533203125,
0.0208892822265625,
-0.038818359375,
0.0045623779296875,
0.01525115966796875,
-0.059600830078125,
-0.050689697265625,
-0.035125732421875,
0.00788116455078125,
0.041717529296875,
-0.0384521484375,
0.043487548828125,
0.0237884521484375,
-0.00940704345703125,
-0.034149169921875,
-0.0830078125,
-0.00711822509765625,
-0.01125335693359375,
-0.0550537109375,
0.03076171875,
-0.01491546630859375,
0.0162353515625,
0.017730712890625,
0.01898193359375,
0.01163482666015625,
-0.0110626220703125,
0.0038166046142578125,
0.02276611328125,
-0.0033397674560546875,
0.0191497802734375,
0.0227203369140625,
-0.01239013671875,
-0.00385284423828125,
-0.0131683349609375,
0.061981201171875,
-0.0239105224609375,
-0.005863189697265625,
-0.02569580078125,
0.01458740234375,
0.0226898193359375,
-0.015655517578125,
0.09124755859375,
0.074462890625,
-0.032867431640625,
-0.0102386474609375,
-0.0311737060546875,
-0.0260772705078125,
-0.03729248046875,
0.034698486328125,
-0.024658203125,
-0.074951171875,
0.0288543701171875,
0.0230712890625,
0.0020046234130859375,
0.057861328125,
0.0498046875,
-0.0242919921875,
0.06707763671875,
0.047698974609375,
-0.0087432861328125,
0.04052734375,
-0.0496826171875,
0.0198211669921875,
-0.06805419921875,
-0.0038604736328125,
-0.025970458984375,
-0.0243988037109375,
-0.04949951171875,
-0.03564453125,
0.01934814453125,
-0.01013946533203125,
-0.01715087890625,
0.047698974609375,
-0.04150390625,
0.01033782958984375,
0.039581298828125,
0.00508880615234375,
0.0008611679077148438,
0.00907135009765625,
-0.04193115234375,
-0.0099945068359375,
-0.06427001953125,
-0.0419921875,
0.053741455078125,
0.0283660888671875,
0.03277587890625,
-0.0092315673828125,
0.051300048828125,
0.01294708251953125,
0.0114593505859375,
-0.055511474609375,
0.042755126953125,
-0.0237884521484375,
-0.033233642578125,
-0.025421142578125,
-0.03057861328125,
-0.07049560546875,
0.0462646484375,
-0.015228271484375,
-0.05267333984375,
0.01018524169921875,
-0.01354217529296875,
-0.027496337890625,
0.0168914794921875,
-0.06036376953125,
0.08160400390625,
0.00897216796875,
0.0018033981323242188,
-0.00499725341796875,
-0.053009033203125,
0.0177764892578125,
0.01313018798828125,
0.0196380615234375,
-0.0080718994140625,
-0.0124359130859375,
0.068603515625,
-0.025787353515625,
0.0699462890625,
-0.009979248046875,
0.0177001953125,
0.0266876220703125,
-0.019866943359375,
0.0309906005859375,
-0.01203155517578125,
-0.0005393028259277344,
0.00836181640625,
-0.0030498504638671875,
-0.032958984375,
-0.034027099609375,
0.052978515625,
-0.0653076171875,
-0.0210723876953125,
-0.04266357421875,
-0.046630859375,
-0.0024738311767578125,
0.019287109375,
0.034210205078125,
0.02545166015625,
-0.0031280517578125,
0.0284423828125,
0.031219482421875,
-0.0206451416015625,
0.056365966796875,
0.0130157470703125,
-0.0103302001953125,
-0.0361328125,
0.043487548828125,
0.01065826416015625,
-0.0021343231201171875,
0.0239410400390625,
0.020172119140625,
-0.0305938720703125,
-0.01113128662109375,
-0.033111572265625,
0.04034423828125,
-0.041656494140625,
-0.0235137939453125,
-0.0775146484375,
-0.038238525390625,
-0.045623779296875,
0.0006918907165527344,
-0.0188140869140625,
-0.028564453125,
-0.03662109375,
-0.0216827392578125,
0.02099609375,
0.032958984375,
-0.0003867149353027344,
0.0308380126953125,
-0.05322265625,
0.01450347900390625,
0.01491546630859375,
-0.0028400421142578125,
-0.0080108642578125,
-0.0657958984375,
-0.03216552734375,
0.0071563720703125,
-0.0236968994140625,
-0.06048583984375,
0.051513671875,
0.0182342529296875,
0.03863525390625,
0.013641357421875,
0.0086212158203125,
0.052978515625,
-0.050018310546875,
0.0631103515625,
0.00499725341796875,
-0.07611083984375,
0.037445068359375,
-0.0003077983856201172,
0.0270843505859375,
0.038665771484375,
0.0203704833984375,
-0.037200927734375,
-0.03131103515625,
-0.05902099609375,
-0.066162109375,
0.05743408203125,
0.041656494140625,
0.034423828125,
-0.0148468017578125,
0.007205963134765625,
-0.0214080810546875,
0.0176544189453125,
-0.07855224609375,
-0.035491943359375,
-0.0182342529296875,
-0.04583740234375,
-0.0291290283203125,
-0.02630615234375,
0.00028014183044433594,
-0.028594970703125,
0.0526123046875,
-0.00337982177734375,
0.064208984375,
0.0211944580078125,
-0.0394287109375,
0.01873779296875,
0.00986480712890625,
0.053375244140625,
0.007843017578125,
-0.00431060791015625,
0.020294189453125,
0.026458740234375,
-0.019500732421875,
-0.00024020671844482422,
0.03369140625,
-0.0100860595703125,
0.0203704833984375,
0.031982421875,
0.0701904296875,
0.0325927734375,
-0.03582763671875,
0.06268310546875,
-0.008453369140625,
-0.00959014892578125,
-0.03466796875,
-0.01007843017578125,
0.018310546875,
0.0206146240234375,
0.0176239013671875,
0.0120849609375,
0.0013914108276367188,
-0.0308380126953125,
0.0261993408203125,
0.01513671875,
-0.0350341796875,
0.0011587142944335938,
0.045928955078125,
-0.0039215087890625,
-0.01071929931640625,
0.06341552734375,
-0.0217132568359375,
-0.049346923828125,
0.033905029296875,
0.046173095703125,
0.0653076171875,
-0.0005879402160644531,
0.016265869140625,
0.032806396484375,
0.03997802734375,
-0.0006799697875976562,
0.0006055831909179688,
0.00815582275390625,
-0.0709228515625,
-0.0113067626953125,
-0.043701171875,
0.01142120361328125,
-0.0100555419921875,
-0.043792724609375,
0.0218963623046875,
-0.004634857177734375,
-0.0060577392578125,
-0.0162811279296875,
0.0032196044921875,
-0.058074951171875,
-0.00029730796813964844,
0.005680084228515625,
0.0693359375,
-0.07904052734375,
0.07427978515625,
0.046356201171875,
-0.054901123046875,
-0.050628662109375,
-0.01531982421875,
-0.0181427001953125,
-0.06585693359375,
0.0262908935546875,
0.0330810546875,
0.007305145263671875,
0.0092010498046875,
-0.04486083984375,
-0.059051513671875,
0.10565185546875,
0.01483154296875,
-0.028411865234375,
-0.0164337158203125,
-0.0011548995971679688,
0.04217529296875,
-0.04541015625,
0.0264739990234375,
0.03411865234375,
0.023193359375,
-0.01311492919921875,
-0.050537109375,
0.01537322998046875,
-0.020751953125,
0.022216796875,
-0.015777587890625,
-0.046539306640625,
0.07080078125,
-0.0064544677734375,
-0.0190582275390625,
0.0193023681640625,
0.070556640625,
0.0303497314453125,
-0.000759124755859375,
0.0389404296875,
0.05364990234375,
0.049530029296875,
-0.0093231201171875,
0.078857421875,
-0.013519287109375,
0.06195068359375,
0.08135986328125,
0.01088714599609375,
0.08050537109375,
0.037200927734375,
-0.0085601806640625,
0.0645751953125,
0.041412353515625,
-0.021026611328125,
0.053253173828125,
0.01110076904296875,
0.005435943603515625,
0.00916290283203125,
0.0076904296875,
-0.01519775390625,
0.0290069580078125,
0.0179443359375,
-0.0540771484375,
0.004917144775390625,
0.0203399658203125,
0.0103759765625,
-0.00365447998046875,
-0.0000029802322387695312,
0.046051025390625,
0.02020263671875,
-0.027069091796875,
0.0294189453125,
0.0182952880859375,
0.07684326171875,
-0.035064697265625,
0.0213623046875,
-0.01561737060546875,
0.0227203369140625,
-0.0042877197265625,
-0.040435791015625,
0.0322265625,
-0.007259368896484375,
-0.00506591796875,
-0.018310546875,
0.052337646484375,
-0.053619384765625,
-0.047576904296875,
0.0239105224609375,
0.03570556640625,
0.0080718994140625,
0.004932403564453125,
-0.07855224609375,
0.005420684814453125,
0.00727081298828125,
-0.03363037109375,
0.0163726806640625,
0.0162506103515625,
0.0279388427734375,
0.041717529296875,
0.03387451171875,
-0.007793426513671875,
0.0187530517578125,
0.00946044921875,
0.061798095703125,
-0.04425048828125,
-0.0413818359375,
-0.0677490234375,
0.054107666015625,
-0.0211639404296875,
-0.0174560546875,
0.06036376953125,
0.042266845703125,
0.0654296875,
-0.0277252197265625,
0.035888671875,
-0.013946533203125,
0.0176239013671875,
-0.0418701171875,
0.0672607421875,
-0.0404052734375,
-0.00217437744140625,
-0.01242828369140625,
-0.06707763671875,
-0.0214691162109375,
0.0826416015625,
-0.0245208740234375,
0.006984710693359375,
0.07763671875,
0.06646728515625,
-0.00677490234375,
-0.00690460205078125,
0.0162353515625,
0.032379150390625,
0.0131683349609375,
0.032928466796875,
0.038787841796875,
-0.0660400390625,
0.058013916015625,
-0.04510498046875,
-0.00516510009765625,
-0.0162200927734375,
-0.054595947265625,
-0.073486328125,
-0.060333251953125,
-0.0232696533203125,
-0.025390625,
-0.0158538818359375,
0.0792236328125,
0.04534912109375,
-0.05999755859375,
-0.01120758056640625,
-0.003498077392578125,
-0.00980377197265625,
-0.01084136962890625,
-0.0245208740234375,
0.0399169921875,
-0.042877197265625,
-0.05902099609375,
0.0173187255859375,
-0.0022716522216796875,
0.003498077392578125,
-0.0217132568359375,
0.006191253662109375,
-0.04364013671875,
0.0163726806640625,
0.053375244140625,
-0.019989013671875,
-0.056732177734375,
-0.0225982666015625,
-0.0014858245849609375,
-0.04205322265625,
-0.007183074951171875,
0.03790283203125,
-0.046600341796875,
0.0223541259765625,
0.0318603515625,
0.035552978515625,
0.0584716796875,
-0.0216827392578125,
0.02215576171875,
-0.06475830078125,
0.0272064208984375,
0.004215240478515625,
0.059844970703125,
0.0259246826171875,
-0.0164337158203125,
0.04583740234375,
0.01947021484375,
-0.03546142578125,
-0.054901123046875,
-0.0099639892578125,
-0.0740966796875,
-0.0164031982421875,
0.08258056640625,
-0.0223388671875,
-0.022918701171875,
0.0146484375,
-0.0263824462890625,
0.032806396484375,
-0.020782470703125,
0.047088623046875,
0.05975341796875,
-0.00734710693359375,
-0.030731201171875,
-0.0290069580078125,
0.02276611328125,
0.03302001953125,
-0.045989990234375,
-0.0227203369140625,
0.01224517822265625,
0.016937255859375,
0.01904296875,
0.0382080078125,
-0.00643157958984375,
-0.00823974609375,
0.004642486572265625,
0.01097869873046875,
-0.01309967041015625,
-0.00649261474609375,
-0.038604736328125,
0.00788116455078125,
-0.031829833984375,
-0.0289306640625
]
] |
bofenghuang/vigogne-2-7b-chat | 2023-10-16T14:03:25.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"LLM",
"llama-2",
"finetuned",
"fr",
"license:llama2",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | bofenghuang | null | null | bofenghuang/vigogne-2-7b-chat | 15 | 7,341 | transformers | 2023-07-29T21:16:01 | ---
license: llama2
language: fr
pipeline_tag: text-generation
inference: false
tags:
- LLM
- llama-2
- finetuned
---
<p align="center" width="100%">
<img src="https://huggingface.co/bofenghuang/vigogne-2-7b-chat/resolve/v2.0/logo_v2.jpg" alt="Vigogne" style="width: 30%; min-width: 300px; display: block; margin: auto;">
</p>
# Vigogne-2-7B-Chat-V2.0: A Llama-2-based French Chat LLM
Vigogne-2-7B-Chat-V2.0 is a French chat LLM, based on [LLaMA-2-7B](https://ai.meta.com/llama), optimized to generate helpful and coherent responses in conversations with users.
Check out our [release blog](https://github.com/bofenghuang/vigogne/blob/main/blogs/2023-08-17-vigogne-chat-v2_0.md) and [GitHub repository](https://github.com/bofenghuang/vigogne) for more information.
**Usage and License Notices**: Vigogne-2-7B-Chat-V2.0 follows Llama-2's [usage policy](https://ai.meta.com/llama/use-policy). A significant portion of the training data is distilled from GPT-3.5-Turbo and GPT-4, kindly use it cautiously to avoid any violations of OpenAI's [terms of use](https://openai.com/policies/terms-of-use).
## Changelog
All previous versions are accessible through branches.
- **V1.0**: Trained on 420K chat data.
- **V2.0**: Trained on 520K data. Check out our [release blog](https://github.com/bofenghuang/vigogne/blob/main/blogs/2023-08-17-vigogne-chat-v2_0.md) for more details.
## Prompt Template
We utilized prefix tokens `<user>:` and `<assistant>:` to distinguish between user and assistant utterances.
You can apply this formatting using the [chat template](https://huggingface.co/docs/transformers/main/chat_templating) through the `apply_chat_template()` method.
```python
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("bofenghuang/vigogne-2-7b-chat")
conversation = [
{"role": "user", "content": "Bonjour ! Comment ça va aujourd'hui ?"},
{"role": "assistant", "content": "Bonjour ! Je suis une IA, donc je n'ai pas de sentiments, mais je suis prêt à vous aider. Comment puis-je vous assister aujourd'hui ?"},
{"role": "user", "content": "Quelle est la hauteur de la Tour Eiffel ?"},
{"role": "assistant", "content": "La Tour Eiffel mesure environ 330 mètres de hauteur."},
{"role": "user", "content": "Comment monter en haut ?"},
]
print(tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True))
```
You will get
```
<s><|system|>: Vous êtes Vigogne, un assistant IA créé par Zaion Lab. Vous suivez extrêmement bien les instructions. Aidez autant que vous le pouvez.
<|user|>: Bonjour ! Comment ça va aujourd'hui ?
<|assistant|>: Bonjour ! Je suis une IA, donc je n'ai pas de sentiments, mais je suis prêt à vous aider. Comment puis-je vous assister aujourd'hui ?</s>
<|user|>: Quelle est la hauteur de la Tour Eiffel ?
<|assistant|>: La Tour Eiffel mesure environ 330 mètres de hauteur.</s>
<|user|>: Comment monter en haut ?
<|assistant|>:
```
## Usage
### Inference using the quantized versions
The quantized versions of this model are generously provided by [TheBloke](https://huggingface.co/TheBloke)!
- AWQ for GPU inference: [TheBloke/Vigogne-2-7B-Chat-AWQ](https://huggingface.co/TheBloke/Vigogne-2-7B-Chat-AWQ)
- GTPQ for GPU inference: [TheBloke/Vigogne-2-7B-Chat-GPTQ](https://huggingface.co/TheBloke/Vigogne-2-7B-Chat-GPTQ)
- GGUF for CPU+GPU inference: [TheBloke/Vigogne-2-7B-Chat-GGUF](https://huggingface.co/TheBloke/Vigogne-2-7B-Chat-GGUF)
These versions facilitate testing and development with various popular frameworks, including [AutoAWQ](https://github.com/casper-hansen/AutoAWQ), [vLLM](https://github.com/vllm-project/vllm), [AutoGPTQ](https://github.com/PanQiWei/AutoGPTQ), [GPTQ-for-LLaMa](https://github.com/qwopqwop200/GPTQ-for-LLaMa), [llama.cpp](https://github.com/ggerganov/llama.cpp), [text-generation-webui](https://github.com/oobabooga/text-generation-webui), and more.
### Inference using the unquantized model with 🤗 Transformers
```python
from typing import Dict, List, Optional
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, GenerationConfig, TextStreamer
model_name_or_path = "bofenghuang/vigogne-2-7b-chat"
revision = "v2.0"
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, revision=revision, padding_side="right", use_fast=False)
model = AutoModelForCausalLM.from_pretrained(model_name_or_path, revision=revision, torch_dtype=torch.float16, device_map="auto")
streamer = TextStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
def chat(
query: str,
history: Optional[List[Dict]] = None,
temperature: float = 0.7,
top_p: float = 1.0,
top_k: float = 0,
repetition_penalty: float = 1.1,
max_new_tokens: int = 1024,
**kwargs,
):
if history is None:
history = []
history.append({"role": "user", "content": query})
input_ids = tokenizer.apply_chat_template(history, add_generation_prompt=True, return_tensors="pt").to(model.device)
input_length = input_ids.shape[1]
generated_outputs = model.generate(
input_ids=input_ids,
generation_config=GenerationConfig(
temperature=temperature,
do_sample=temperature > 0.0,
top_p=top_p,
top_k=top_k,
repetition_penalty=repetition_penalty,
max_new_tokens=max_new_tokens,
pad_token_id=tokenizer.eos_token_id,
**kwargs,
),
streamer=streamer,
return_dict_in_generate=True,
)
generated_tokens = generated_outputs.sequences[0, input_length:]
generated_text = tokenizer.decode(generated_tokens, skip_special_tokens=True)
history.append({"role": "assistant", "content": generated_text})
return generated_text, history
# 1st round
response, history = chat("Un escargot parcourt 100 mètres en 5 heures. Quelle est sa vitesse ?", history=None)
# 2nd round
response, history = chat("Quand il peut dépasser le lapin ?", history=history)
# 3rd round
response, history = chat("Écris une histoire imaginative qui met en scène une compétition de course entre un escargot et un lapin.", history=history)
```
You can also use the Google Colab Notebook provided below.
<a href="https://colab.research.google.com/github/bofenghuang/vigogne/blob/main/notebooks/infer_chat.ipynb" target="_blank"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/></a>
### Inference using the unquantized model with vLLM
Set up an OpenAI-compatible server with the following command:
```bash
# Install vLLM
# This may take 5-10 minutes.
# pip install vllm
# Start server for Vigogne-Chat models
python -m vllm.entrypoints.openai.api_server --model bofenghuang/vigogne-2-7b-chat
# List models
# curl http://localhost:8000/v1/models
```
Query the model using the openai python package.
```python
import openai
# Modify OpenAI's API key and API base to use vLLM's API server.
openai.api_key = "EMPTY"
openai.api_base = "http://localhost:8000/v1"
# First model
models = openai.Model.list()
model = models["data"][0]["id"]
# Chat completion API
chat_completion = openai.ChatCompletion.create(
model=model,
messages=[
{"role": "user", "content": "Parle-moi de toi-même."},
],
max_tokens=1024,
temperature=0.7,
)
print("Chat completion results:", chat_completion)
```
## Limitations
Vigogne is still under development, and there are many limitations that have to be addressed. Please note that it is possible that the model generates harmful or biased content, incorrect information or generally unhelpful answers.
| 7,630 | [
[
-0.022003173828125,
-0.069580078125,
0.014617919921875,
0.0276336669921875,
-0.0157928466796875,
-0.01154327392578125,
-0.0214691162109375,
-0.025634765625,
0.0204620361328125,
0.0132293701171875,
-0.04119873046875,
-0.043914794921875,
-0.03289794921875,
-0.006298065185546875,
-0.013427734375,
0.0576171875,
0.01383209228515625,
-0.0169219970703125,
-0.01995849609375,
-0.0039215087890625,
-0.0276336669921875,
-0.045257568359375,
-0.06451416015625,
-0.02099609375,
0.007007598876953125,
0.01377105712890625,
0.0274658203125,
0.0189208984375,
0.0128936767578125,
0.0338134765625,
-0.0179443359375,
0.036163330078125,
-0.047515869140625,
0.003978729248046875,
0.00928497314453125,
-0.03643798828125,
-0.052276611328125,
-0.004535675048828125,
0.0252838134765625,
0.00333404541015625,
-0.0078887939453125,
0.015869140625,
0.024871826171875,
0.01244354248046875,
-0.03314208984375,
0.028656005859375,
-0.035125732421875,
-0.0175628662109375,
0.003421783447265625,
-0.01256561279296875,
-0.028106689453125,
-0.003391265869140625,
0.0109710693359375,
-0.052520751953125,
0.00678253173828125,
-0.0065460205078125,
0.09979248046875,
0.01122283935546875,
-0.0295257568359375,
-0.012054443359375,
-0.040191650390625,
0.0684814453125,
-0.07177734375,
0.013702392578125,
0.036224365234375,
0.00946044921875,
-0.03125,
-0.07049560546875,
-0.052215576171875,
-0.01219940185546875,
-0.018829345703125,
0.024749755859375,
-0.0340576171875,
-0.006031036376953125,
0.00630950927734375,
0.023895263671875,
-0.05126953125,
0.002620697021484375,
-0.04052734375,
-0.0190582275390625,
0.06341552734375,
0.0160980224609375,
0.025299072265625,
-0.0303192138671875,
-0.024505615234375,
-0.0190582275390625,
-0.0211029052734375,
0.0152587890625,
0.0218658447265625,
0.0102996826171875,
-0.04901123046875,
0.0343017578125,
-0.035430908203125,
0.046051025390625,
0.009674072265625,
-0.0159759521484375,
0.034576416015625,
-0.0276641845703125,
-0.03314208984375,
-0.02880859375,
0.10882568359375,
0.038787841796875,
0.0006704330444335938,
0.01336669921875,
0.0148773193359375,
-0.01319122314453125,
-0.0120849609375,
-0.05841064453125,
-0.0229644775390625,
0.03778076171875,
-0.0269622802734375,
-0.035247802734375,
-0.00363922119140625,
-0.0477294921875,
-0.00791168212890625,
0.0112457275390625,
0.034820556640625,
-0.0248260498046875,
-0.032012939453125,
0.0067291259765625,
-0.0135498046875,
0.033905029296875,
0.0223388671875,
-0.061126708984375,
0.004550933837890625,
0.0189056396484375,
0.0662841796875,
0.0168914794921875,
-0.0145111083984375,
-0.019256591796875,
-0.011016845703125,
0.0010356903076171875,
0.046783447265625,
-0.0124664306640625,
-0.041412353515625,
-0.015899658203125,
0.01531982421875,
-0.0139312744140625,
-0.014617919921875,
0.03594970703125,
-0.015838623046875,
0.047119140625,
-0.00511932373046875,
-0.033935546875,
-0.01617431640625,
0.015777587890625,
-0.0200653076171875,
0.0870361328125,
0.0016946792602539062,
-0.061279296875,
-0.0011930465698242188,
-0.039215087890625,
-0.0156402587890625,
-0.0105438232421875,
-0.007221221923828125,
-0.03924560546875,
-0.020599365234375,
0.037689208984375,
0.0367431640625,
-0.03070068359375,
-0.0023651123046875,
-0.0304107666015625,
-0.0174560546875,
0.02734375,
-0.03558349609375,
0.09332275390625,
0.0155487060546875,
-0.03045654296875,
0.0187530517578125,
-0.046875,
0.0111083984375,
0.03167724609375,
-0.01084136962890625,
-0.01030731201171875,
0.0005822181701660156,
0.0010786056518554688,
0.0163116455078125,
0.023529052734375,
-0.028472900390625,
0.0247802734375,
-0.03192138671875,
0.05841064453125,
0.04974365234375,
0.00370025634765625,
0.0311431884765625,
-0.025238037109375,
0.01971435546875,
0.007411956787109375,
0.037628173828125,
-0.0027904510498046875,
-0.05224609375,
-0.07501220703125,
-0.027313232421875,
0.0113372802734375,
0.06243896484375,
-0.054718017578125,
0.05474853515625,
-0.0167236328125,
-0.0552978515625,
-0.044830322265625,
0.005290985107421875,
0.022247314453125,
0.0279998779296875,
0.0240325927734375,
-0.01363372802734375,
-0.041412353515625,
-0.045257568359375,
0.00412750244140625,
-0.032562255859375,
-0.0127105712890625,
0.045867919921875,
0.04107666015625,
-0.021759033203125,
0.0596923828125,
-0.044189453125,
-0.01568603515625,
-0.015228271484375,
0.00010281801223754883,
0.0212860107421875,
0.047882080078125,
0.0478515625,
-0.05157470703125,
-0.037750244140625,
-0.00841522216796875,
-0.07452392578125,
0.0003886222839355469,
-0.007114410400390625,
-0.03668212890625,
0.01490020751953125,
0.0288543701171875,
-0.058197021484375,
0.045562744140625,
0.04290771484375,
-0.038665771484375,
0.019256591796875,
-0.0173187255859375,
0.0182342529296875,
-0.10009765625,
-0.00803375244140625,
0.007556915283203125,
-0.022216796875,
-0.04486083984375,
-0.0145111083984375,
-0.01480865478515625,
0.0159759521484375,
-0.0416259765625,
0.06170654296875,
-0.03692626953125,
0.0307159423828125,
-0.005970001220703125,
0.0234375,
0.00905609130859375,
0.047882080078125,
-0.0015344619750976562,
0.044097900390625,
0.057281494140625,
-0.045501708984375,
0.044708251953125,
0.042510986328125,
-0.0034809112548828125,
0.0263671875,
-0.07257080078125,
0.029510498046875,
0.0023670196533203125,
0.0207366943359375,
-0.0938720703125,
-0.0222625732421875,
0.057830810546875,
-0.07000732421875,
0.014984130859375,
-0.0323486328125,
-0.032745361328125,
-0.03546142578125,
-0.0191192626953125,
0.01116943359375,
0.05889892578125,
-0.03167724609375,
0.0509033203125,
0.0229339599609375,
0.005084991455078125,
-0.05194091796875,
-0.052734375,
-0.005001068115234375,
-0.026580810546875,
-0.06878662109375,
0.0134735107421875,
-0.0015430450439453125,
-0.0027942657470703125,
-0.00717926025390625,
0.00852203369140625,
-0.003154754638671875,
0.0090484619140625,
0.02606201171875,
0.0295562744140625,
-0.004978179931640625,
-0.01226043701171875,
-0.005794525146484375,
-0.003284454345703125,
-0.0012998580932617188,
-0.02410888671875,
0.057952880859375,
-0.0274505615234375,
-0.0080108642578125,
-0.050750732421875,
0.00836181640625,
0.03167724609375,
-0.01543426513671875,
0.07928466796875,
0.0625,
-0.0188140869140625,
0.00576019287109375,
-0.04119873046875,
-0.0186920166015625,
-0.040496826171875,
0.0193634033203125,
-0.020416259765625,
-0.056793212890625,
0.056671142578125,
0.036041259765625,
0.01953125,
0.050750732421875,
0.049285888671875,
-0.0108642578125,
0.0736083984375,
0.0300750732421875,
-0.003253936767578125,
0.041473388671875,
-0.052001953125,
0.00571441650390625,
-0.054107666015625,
-0.027313232421875,
-0.0243072509765625,
-0.00040984153747558594,
-0.05657958984375,
-0.040679931640625,
0.0225067138671875,
0.019195556640625,
-0.0179901123046875,
0.03045654296875,
-0.04705810546875,
0.003082275390625,
0.050994873046875,
0.0214691162109375,
0.006572723388671875,
-0.0024089813232421875,
0.00811004638671875,
0.0127105712890625,
-0.051727294921875,
-0.03887939453125,
0.0802001953125,
0.0249481201171875,
0.056060791015625,
0.0010738372802734375,
0.049774169921875,
-0.00389862060546875,
0.0170745849609375,
-0.0352783203125,
0.04608154296875,
0.0179901123046875,
-0.0389404296875,
-0.0269927978515625,
-0.038330078125,
-0.06927490234375,
0.0307464599609375,
-0.019744873046875,
-0.07720947265625,
0.0184326171875,
0.0130615234375,
-0.0237884521484375,
0.01409912109375,
-0.05535888671875,
0.06939697265625,
-0.00800323486328125,
-0.0301666259765625,
0.00640106201171875,
-0.042327880859375,
0.024169921875,
0.032135009765625,
0.003215789794921875,
-0.0149383544921875,
0.00736236572265625,
0.056488037109375,
-0.041534423828125,
0.07122802734375,
-0.006206512451171875,
0.00388336181640625,
0.049774169921875,
-0.01146697998046875,
0.037017822265625,
0.0244903564453125,
0.00965118408203125,
0.029205322265625,
-0.00043892860412597656,
-0.036529541015625,
-0.037750244140625,
0.054595947265625,
-0.077392578125,
-0.04827880859375,
-0.032745361328125,
-0.022003173828125,
0.0111846923828125,
0.0091705322265625,
0.053070068359375,
0.027130126953125,
-0.003864288330078125,
0.00812530517578125,
0.03668212890625,
-0.034820556640625,
0.031402587890625,
0.0218658447265625,
-0.016021728515625,
-0.039947509765625,
0.056854248046875,
-0.007717132568359375,
0.020965576171875,
0.0127716064453125,
0.00257110595703125,
-0.0313720703125,
-0.024139404296875,
-0.035003662109375,
0.0261993408203125,
-0.04498291015625,
-0.019287109375,
-0.068603515625,
-0.031158447265625,
-0.054901123046875,
0.00945281982421875,
-0.031219482421875,
-0.018035888671875,
-0.05242919921875,
-0.0017948150634765625,
0.054443359375,
0.017120361328125,
-0.0037631988525390625,
0.0283660888671875,
-0.046417236328125,
0.01593017578125,
0.0253753662109375,
-0.0004131793975830078,
-0.007724761962890625,
-0.0609130859375,
-0.005672454833984375,
0.0318603515625,
-0.031768798828125,
-0.053131103515625,
0.04302978515625,
0.00820159912109375,
0.04400634765625,
0.0231781005859375,
0.01039886474609375,
0.058746337890625,
-0.005359649658203125,
0.07269287109375,
0.009674072265625,
-0.06390380859375,
0.0458984375,
-0.031341552734375,
0.0152130126953125,
0.01386260986328125,
0.0202484130859375,
-0.04229736328125,
-0.03668212890625,
-0.059417724609375,
-0.07476806640625,
0.0518798828125,
0.050079345703125,
0.025726318359375,
-0.0094451904296875,
0.01995849609375,
-0.03192138671875,
0.00004410743713378906,
-0.05975341796875,
-0.049163818359375,
-0.0304107666015625,
-0.01088714599609375,
0.004573822021484375,
-0.00931549072265625,
-0.00432586669921875,
-0.033782958984375,
0.0621337890625,
-0.0027942657470703125,
0.05615234375,
0.038909912109375,
-0.00800323486328125,
0.01470947265625,
0.00727081298828125,
0.0489501953125,
0.039306640625,
-0.00948333740234375,
-0.0207061767578125,
0.0305633544921875,
-0.0257720947265625,
0.0017004013061523438,
0.016693115234375,
-0.0106353759765625,
0.0016164779663085938,
0.0230560302734375,
0.07965087890625,
-0.00513458251953125,
-0.02825927734375,
0.0484619140625,
-0.02972412109375,
-0.0260009765625,
-0.03973388671875,
0.0167388916015625,
0.021240234375,
0.034698486328125,
0.026031494140625,
-0.0175933837890625,
-0.0032672882080078125,
-0.032623291015625,
-0.00043463706970214844,
0.0310516357421875,
-0.01837158203125,
-0.0177154541015625,
0.0845947265625,
0.015228271484375,
-0.0134124755859375,
0.057220458984375,
-0.02117919921875,
-0.032196044921875,
0.04766845703125,
0.0252838134765625,
0.06219482421875,
-0.003589630126953125,
0.013671875,
0.0457763671875,
0.0106201171875,
-0.00830841064453125,
0.0248260498046875,
0.0037078857421875,
-0.0489501953125,
-0.0250244140625,
-0.038116455078125,
-0.0022830963134765625,
0.0235137939453125,
-0.041015625,
0.0297393798828125,
-0.02294921875,
-0.031158447265625,
-0.00589752197265625,
-0.00524139404296875,
-0.057159423828125,
0.00353240966796875,
0.007587432861328125,
0.049468994140625,
-0.0506591796875,
0.047576904296875,
0.044647216796875,
-0.04449462890625,
-0.06170654296875,
-0.01335906982421875,
0.00839996337890625,
-0.0545654296875,
0.022430419921875,
0.004070281982421875,
0.005695343017578125,
0.005222320556640625,
-0.057525634765625,
-0.056884765625,
0.0972900390625,
0.01355743408203125,
-0.0323486328125,
-0.01538848876953125,
-0.00852203369140625,
0.040771484375,
-0.016632080078125,
0.05859375,
0.0300140380859375,
0.017181396484375,
0.0134735107421875,
-0.087158203125,
0.011383056640625,
-0.0355224609375,
-0.0011234283447265625,
-0.0092010498046875,
-0.08074951171875,
0.06787109375,
-0.0182037353515625,
-0.01678466796875,
0.016265869140625,
0.08099365234375,
0.0126953125,
0.00921630859375,
0.02825927734375,
0.031646728515625,
0.05126953125,
-0.020538330078125,
0.06976318359375,
-0.0189208984375,
0.0621337890625,
0.06103515625,
0.01220703125,
0.050445556640625,
0.0126953125,
-0.01549530029296875,
0.03814697265625,
0.054962158203125,
-0.00786590576171875,
0.0300750732421875,
0.0007143020629882812,
-0.0226287841796875,
-0.018096923828125,
0.012420654296875,
-0.034912109375,
0.040374755859375,
0.0247650146484375,
-0.01641845703125,
0.00003457069396972656,
0.001800537109375,
0.022705078125,
-0.024322509765625,
0.004878997802734375,
0.057769775390625,
0.016693115234375,
-0.03533935546875,
0.0836181640625,
-0.0003902912139892578,
0.0689697265625,
-0.03515625,
0.0007224082946777344,
-0.0217742919921875,
0.01148223876953125,
-0.0167236328125,
-0.04766845703125,
-0.0019969940185546875,
-0.005298614501953125,
0.01220703125,
0.00528717041015625,
0.05059814453125,
-0.034881591796875,
-0.0288543701171875,
0.022064208984375,
0.033355712890625,
0.0248260498046875,
0.0004112720489501953,
-0.0631103515625,
0.009674072265625,
0.01168060302734375,
-0.049163818359375,
0.0088653564453125,
0.0209503173828125,
0.005252838134765625,
0.058563232421875,
0.046539306640625,
-0.01309967041015625,
0.0031452178955078125,
-0.01197052001953125,
0.06475830078125,
-0.04107666015625,
-0.033477783203125,
-0.0743408203125,
0.061737060546875,
-0.025848388671875,
-0.035247802734375,
0.06475830078125,
0.033660888671875,
0.06610107421875,
0.003658294677734375,
0.0653076171875,
-0.029998779296875,
0.01042938232421875,
-0.0286102294921875,
0.057220458984375,
-0.045806884765625,
0.017852783203125,
-0.0192108154296875,
-0.040863037109375,
0.0025997161865234375,
0.06103515625,
-0.0195465087890625,
0.0088958740234375,
0.0263671875,
0.0755615234375,
-0.0016155242919921875,
-0.01464080810546875,
0.01288604736328125,
0.0299530029296875,
0.0369873046875,
0.052703857421875,
0.0550537109375,
-0.06787109375,
0.05999755859375,
-0.031646728515625,
-0.0184326171875,
-0.0263824462890625,
-0.05059814453125,
-0.09808349609375,
-0.0574951171875,
-0.030853271484375,
-0.0645751953125,
-0.0019683837890625,
0.07476806640625,
0.05804443359375,
-0.046783447265625,
-0.0149993896484375,
0.0060272216796875,
0.00678253173828125,
-0.01291656494140625,
-0.0214996337890625,
0.027252197265625,
-0.00771331787109375,
-0.0743408203125,
0.0106201171875,
-0.005657196044921875,
0.0296630859375,
-0.00205230712890625,
-0.0098876953125,
-0.0162811279296875,
0.00875091552734375,
0.04095458984375,
0.038116455078125,
-0.056060791015625,
-0.01528167724609375,
0.005710601806640625,
-0.026641845703125,
0.011260986328125,
0.01401519775390625,
-0.03985595703125,
0.022369384765625,
0.04949951171875,
0.004024505615234375,
0.059326171875,
-0.004207611083984375,
0.03375244140625,
-0.04083251953125,
0.0406494140625,
-0.006092071533203125,
0.032440185546875,
0.019805908203125,
-0.03533935546875,
0.034637451171875,
0.02239990234375,
-0.04608154296875,
-0.063232421875,
-0.0161285400390625,
-0.0860595703125,
-0.0095062255859375,
0.101806640625,
-0.0183868408203125,
-0.02606201171875,
0.004062652587890625,
-0.034454345703125,
0.037628173828125,
-0.037750244140625,
0.047119140625,
0.040374755859375,
-0.0126953125,
-0.0174102783203125,
-0.049896240234375,
0.034942626953125,
0.0149383544921875,
-0.06072998046875,
-0.006488800048828125,
0.021728515625,
0.045318603515625,
-0.0001163482666015625,
0.0738525390625,
0.006683349609375,
0.0162506103515625,
-0.000370025634765625,
0.00937652587890625,
-0.00926971435546875,
0.0028438568115234375,
-0.0035076141357421875,
-0.0232391357421875,
-0.019195556640625,
-0.0309906005859375
]
] |
Writer/palmyra-20b-chat | 2023-08-28T17:46:10.000Z | [
"transformers",
"pytorch",
"gpt2",
"text-generation",
"chat",
"palmyra",
"en",
"dataset:WizardLM/WizardLM_evol_instruct_V2_196k",
"dataset:Open-Orca/OpenOrca",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | Writer | null | null | Writer/palmyra-20b-chat | 5 | 7,337 | transformers | 2023-08-23T18:29:54 | ---
datasets:
- WizardLM/WizardLM_evol_instruct_V2_196k
- Open-Orca/OpenOrca
language:
- en
tags:
- chat
- palmyra
---
# Writer/palmyra-20b-chat
---
# Usage
```py
import torch
from transformers import AutoTokenizer, AutoModelForCausalLM, TextStreamer
model_name = "Writer/palmyra-20b-chat"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
model_name,
torch_dtype=torch.float16,
device_map="auto",
)
prompt = "What is the meaning of life?"
input_text = (
"A chat between a curious user and an artificial intelligence assistant. "
"The assistant gives helpful, detailed, and polite answers to the user's questions. "
"USER: {prompt} "
"ASSISTANT:"
)
model_inputs = tokenizer(input_text.format(prompt=prompt), return_tensors="pt").to(
"cuda"
)
gen_conf = {
"top_k": 20,
"max_new_tokens": 2048,
"temperature": 0.6,
"do_sample": True,
"eos_token_id": tokenizer.eos_token_id,
}
streamer = TextStreamer(tokenizer)
if "token_type_ids" in model_inputs:
del model_inputs["token_type_ids"]
all_inputs = {**model_inputs, **gen_conf}
output = model.generate(**all_inputs, streamer=streamer)
print("-"*20)
print(output)
``` | 1,231 | [
[
-0.0225372314453125,
-0.0538330078125,
0.0103759765625,
0.020751953125,
-0.018798828125,
0.0127716064453125,
-0.0092926025390625,
-0.01378631591796875,
0.00273895263671875,
0.0248260498046875,
-0.034027099609375,
-0.032989501953125,
-0.061370849609375,
-0.0012950897216796875,
-0.038177490234375,
0.09405517578125,
-0.009033203125,
-0.025421142578125,
0.01415252685546875,
0.00804901123046875,
-0.024017333984375,
-0.0300140380859375,
-0.08349609375,
-0.01593017578125,
0.0179290771484375,
0.004802703857421875,
0.04644775390625,
0.037994384765625,
0.018218994140625,
0.0361328125,
-0.005001068115234375,
0.0249481201171875,
-0.0254669189453125,
0.01088714599609375,
0.0076446533203125,
-0.029754638671875,
-0.033843994140625,
0.00415802001953125,
0.036041259765625,
0.0157012939453125,
0.0126495361328125,
0.03326416015625,
0.0087890625,
0.0007510185241699219,
-0.04791259765625,
0.021270751953125,
-0.0404052734375,
-0.011199951171875,
-0.0080718994140625,
-0.0255889892578125,
-0.03765869140625,
-0.01171112060546875,
0.004924774169921875,
-0.0506591796875,
0.033111572265625,
-0.00037479400634765625,
0.09771728515625,
0.03131103515625,
-0.032379150390625,
-0.036407470703125,
-0.0665283203125,
0.047882080078125,
-0.061248779296875,
-0.0092315673828125,
0.0128021240234375,
0.016448974609375,
-0.0048828125,
-0.0733642578125,
-0.060302734375,
-0.005641937255859375,
-0.006763458251953125,
0.00601959228515625,
-0.0316162109375,
0.00704193115234375,
0.01253509521484375,
0.010955810546875,
-0.03424072265625,
-0.00803375244140625,
-0.054901123046875,
-0.0321044921875,
0.04608154296875,
0.035491943359375,
0.0188140869140625,
-0.01033782958984375,
-0.0208892822265625,
-0.0186309814453125,
-0.012481689453125,
0.004810333251953125,
0.040374755859375,
0.01076507568359375,
-0.0209503173828125,
0.04296875,
-0.00452423095703125,
0.04412841796875,
0.04119873046875,
-0.0023956298828125,
0.02276611328125,
-0.01538848876953125,
-0.0299530029296875,
-0.0043487548828125,
0.07720947265625,
0.0250244140625,
0.00601959228515625,
0.00482177734375,
0.00572967529296875,
0.004253387451171875,
0.0029144287109375,
-0.06982421875,
-0.04730224609375,
0.03662109375,
-0.04510498046875,
-0.042388916015625,
0.00449371337890625,
-0.0413818359375,
-0.01531219482421875,
0.008026123046875,
0.044219970703125,
-0.02197265625,
-0.02972412109375,
-0.01605224609375,
-0.0235137939453125,
0.0232086181640625,
-0.00397491455078125,
-0.09661865234375,
0.00997161865234375,
0.017974853515625,
0.047637939453125,
0.0012655258178710938,
-0.035491943359375,
-0.0189666748046875,
0.013336181640625,
0.003902435302734375,
0.04144287109375,
-0.0067291259765625,
-0.043060302734375,
-0.018157958984375,
0.01136016845703125,
-0.0230865478515625,
-0.031951904296875,
0.029754638671875,
-0.023468017578125,
0.030975341796875,
-0.007282257080078125,
-0.042816162109375,
-0.00916290283203125,
0.004779815673828125,
-0.01763916015625,
0.07525634765625,
0.00397491455078125,
-0.0594482421875,
0.0301666259765625,
-0.05242919921875,
-0.016021728515625,
0.00838470458984375,
-0.0119476318359375,
-0.047088623046875,
-0.0004525184631347656,
0.018096923828125,
0.031402587890625,
0.00811767578125,
0.0271453857421875,
-0.018035888671875,
-0.0181427001953125,
0.01983642578125,
-0.054473876953125,
0.072998046875,
0.0275726318359375,
-0.04937744140625,
0.051422119140625,
-0.037109375,
-0.01082611083984375,
0.0066070556640625,
-0.0206451416015625,
0.0111541748046875,
-0.01318359375,
0.038665771484375,
0.00980377197265625,
0.031097412109375,
-0.056640625,
0.021209716796875,
-0.04107666015625,
0.051849365234375,
0.05303955078125,
-0.004337310791015625,
0.038330078125,
-0.02740478515625,
0.0182037353515625,
0.019256591796875,
0.021240234375,
0.0059661865234375,
-0.0278778076171875,
-0.087890625,
-0.013916015625,
-0.00337982177734375,
0.035430908203125,
-0.060821533203125,
0.055419921875,
-0.0031833648681640625,
-0.059234619140625,
-0.031585693359375,
-0.012481689453125,
0.00460052490234375,
0.041961669921875,
0.02276611328125,
0.00921630859375,
-0.05218505859375,
-0.047119140625,
-0.0131072998046875,
-0.01047515869140625,
0.005237579345703125,
0.01120758056640625,
0.053802490234375,
-0.021209716796875,
0.068603515625,
-0.042999267578125,
-0.015899658203125,
-0.03265380859375,
0.02349853515625,
0.03741455078125,
0.057525634765625,
0.043487548828125,
-0.035003662109375,
-0.039794921875,
-0.0247650146484375,
-0.04754638671875,
0.007152557373046875,
-0.033294677734375,
-0.0233001708984375,
0.0206451416015625,
0.01253509521484375,
-0.06597900390625,
0.0489501953125,
0.031982421875,
-0.057464599609375,
0.043701171875,
-0.035247802734375,
0.016265869140625,
-0.09722900390625,
0.01092529296875,
-0.0274658203125,
0.0025157928466796875,
-0.0321044921875,
-0.0124664306640625,
-0.01132965087890625,
-0.005218505859375,
-0.044525146484375,
0.06317138671875,
-0.0202484130859375,
0.012054443359375,
-0.04217529296875,
-0.0025348663330078125,
-0.00601959228515625,
0.034027099609375,
-0.0023937225341796875,
0.047454833984375,
0.056396484375,
-0.05950927734375,
0.059234619140625,
0.040924072265625,
0.0171661376953125,
0.004253387451171875,
-0.056488037109375,
0.0012903213500976562,
0.0013256072998046875,
0.0245208740234375,
-0.09161376953125,
-0.00909423828125,
0.0325927734375,
-0.064697265625,
0.0142974853515625,
0.00616455078125,
-0.026641845703125,
-0.0390625,
-0.005077362060546875,
0.032379150390625,
0.0250244140625,
-0.035797119140625,
0.06927490234375,
0.0243988037109375,
0.009307861328125,
-0.040924072265625,
-0.047088623046875,
-0.003917694091796875,
-0.0186767578125,
-0.045379638671875,
0.0325927734375,
-0.009796142578125,
0.0133056640625,
0.0011034011840820312,
0.0024871826171875,
-0.005138397216796875,
0.0036487579345703125,
0.0291290283203125,
0.044036865234375,
0.00347900390625,
-0.007724761962890625,
0.006465911865234375,
-0.027679443359375,
0.029083251953125,
-0.011199951171875,
0.06854248046875,
-0.0133209228515625,
-0.016326904296875,
-0.045013427734375,
0.00984954833984375,
0.034576416015625,
-0.0036163330078125,
0.072265625,
0.0791015625,
-0.0251922607421875,
-0.00009328126907348633,
-0.0294342041015625,
-0.0207977294921875,
-0.0389404296875,
0.031402587890625,
-0.028839111328125,
-0.03851318359375,
0.06427001953125,
0.0262451171875,
0.0333251953125,
0.060089111328125,
0.0660400390625,
-0.01293182373046875,
0.07763671875,
0.031646728515625,
0.0017652511596679688,
0.031463623046875,
-0.057861328125,
0.0258026123046875,
-0.051788330078125,
-0.0186309814453125,
-0.030242919921875,
-0.01629638671875,
-0.046234130859375,
0.0007700920104980469,
0.019287109375,
-0.004924774169921875,
-0.038330078125,
0.045928955078125,
-0.0655517578125,
0.00843048095703125,
0.050384521484375,
0.01088714599609375,
-0.0011339187622070312,
-0.0278167724609375,
-0.0075225830078125,
0.016326904296875,
-0.046783447265625,
-0.0430908203125,
0.0887451171875,
0.012420654296875,
0.035614013671875,
-0.0034046173095703125,
0.065185546875,
-0.009033203125,
0.0309295654296875,
-0.04083251953125,
0.030975341796875,
0.005115509033203125,
-0.07659912109375,
-0.0210723876953125,
-0.0186309814453125,
-0.071533203125,
0.01312255859375,
-0.010498046875,
-0.05853271484375,
0.00508880615234375,
0.0245361328125,
-0.05096435546875,
0.0469970703125,
-0.031646728515625,
0.07757568359375,
-0.005352020263671875,
-0.01617431640625,
0.00296783447265625,
-0.030670166015625,
0.0245208740234375,
0.0269317626953125,
-0.004749298095703125,
-0.0083770751953125,
0.007190704345703125,
0.08056640625,
-0.033660888671875,
0.04742431640625,
-0.0159759521484375,
0.017974853515625,
0.03643798828125,
-0.003448486328125,
0.0218658447265625,
0.00975799560546875,
0.0133056640625,
-0.0098876953125,
0.028778076171875,
-0.034149169921875,
-0.027679443359375,
0.04986572265625,
-0.09033203125,
-0.038665771484375,
-0.04150390625,
-0.040863037109375,
0.01357269287109375,
0.004913330078125,
0.061065673828125,
0.0369873046875,
0.00595855712890625,
0.0012722015380859375,
0.0284576416015625,
-0.0188751220703125,
0.07684326171875,
-0.002552032470703125,
-0.0167999267578125,
-0.0347900390625,
0.045013427734375,
-0.006198883056640625,
0.011688232421875,
0.00495147705078125,
-0.0087738037109375,
-0.0304412841796875,
-0.0262451171875,
-0.048248291015625,
0.0114898681640625,
-0.056396484375,
-0.0189971923828125,
-0.06219482421875,
-0.0347900390625,
-0.038970947265625,
-0.00693511962890625,
-0.030853271484375,
-0.00994110107421875,
-0.05877685546875,
-0.0002256631851196289,
0.038604736328125,
0.02227783203125,
-0.0008020401000976562,
0.047149658203125,
-0.06805419921875,
0.016632080078125,
0.0128936767578125,
-0.00968170166015625,
0.0127716064453125,
-0.06256103515625,
-0.007595062255859375,
0.004302978515625,
-0.047271728515625,
-0.07196044921875,
0.0655517578125,
-0.0026187896728515625,
0.042144775390625,
0.02752685546875,
0.0103302001953125,
0.05828857421875,
0.00469207763671875,
0.053619384765625,
0.0011234283447265625,
-0.091796875,
0.0311737060546875,
-0.0302886962890625,
0.028778076171875,
0.0255889892578125,
0.00824737548828125,
-0.036163330078125,
-0.03277587890625,
-0.0657958984375,
-0.0791015625,
0.039398193359375,
0.034454345703125,
0.0352783203125,
-0.002620697021484375,
0.01513671875,
-0.003917694091796875,
0.004322052001953125,
-0.06988525390625,
-0.04949951171875,
-0.0243988037109375,
-0.033538818359375,
0.0186920166015625,
0.007610321044921875,
0.0083160400390625,
-0.039886474609375,
0.06610107421875,
0.004344940185546875,
0.045013427734375,
0.012603759765625,
-0.00628662109375,
-0.0084228515625,
0.005069732666015625,
0.044525146484375,
0.0484619140625,
-0.009307861328125,
0.000050067901611328125,
0.02984619140625,
-0.037994384765625,
0.004123687744140625,
0.00665283203125,
-0.0068206787109375,
0.00019121170043945312,
0.020751953125,
0.0684814453125,
-0.00313568115234375,
-0.01229095458984375,
0.021759033203125,
-0.0087432861328125,
-0.0213165283203125,
-0.053497314453125,
0.016387939453125,
0.0201568603515625,
0.0150146484375,
0.048065185546875,
0.02740478515625,
-0.005016326904296875,
-0.0278778076171875,
0.005283355712890625,
0.02215576171875,
-0.00725555419921875,
-0.00785064697265625,
0.079833984375,
0.029022216796875,
-0.0180816650390625,
0.0595703125,
-0.00798797607421875,
-0.037109375,
0.058258056640625,
0.0258941650390625,
0.0626220703125,
0.00970458984375,
0.00112152099609375,
0.03411865234375,
0.0074462890625,
0.0256195068359375,
0.023223876953125,
-0.00970458984375,
-0.06402587890625,
-0.01507568359375,
-0.045684814453125,
-0.01079559326171875,
0.01515960693359375,
-0.03436279296875,
0.0271453857421875,
-0.0289306640625,
-0.0302886962890625,
0.0151824951171875,
0.012176513671875,
-0.056396484375,
0.006656646728515625,
0.004680633544921875,
0.0452880859375,
-0.0714111328125,
0.06787109375,
0.0247650146484375,
-0.04632568359375,
-0.09014892578125,
-0.03338623046875,
-0.01763916015625,
-0.043548583984375,
0.0550537109375,
0.00725555419921875,
0.0161895751953125,
0.03173828125,
-0.0323486328125,
-0.0723876953125,
0.086181640625,
-0.0018281936645507812,
-0.0238189697265625,
-0.018585205078125,
0.0005102157592773438,
0.0258636474609375,
-0.02923583984375,
0.0616455078125,
0.03741455078125,
0.039276123046875,
0.01055908203125,
-0.0732421875,
0.005764007568359375,
-0.0225982666015625,
-0.01375579833984375,
0.01373291015625,
-0.037750244140625,
0.09722900390625,
-0.01139068603515625,
-0.009124755859375,
0.0242462158203125,
0.068359375,
0.0377197265625,
0.01348876953125,
0.0205841064453125,
0.0270843505859375,
0.02655029296875,
-0.0269012451171875,
0.04437255859375,
-0.03509521484375,
0.0653076171875,
0.05633544921875,
0.006198883056640625,
0.044219970703125,
0.03436279296875,
-0.0037841796875,
0.038421630859375,
0.06268310546875,
-0.03326416015625,
0.0287322998046875,
0.01739501953125,
-0.0208740234375,
-0.006969451904296875,
0.0305328369140625,
-0.0161285400390625,
0.03857421875,
0.01727294921875,
-0.043731689453125,
-0.0195159912109375,
0.0244293212890625,
0.022491455078125,
-0.023284912109375,
-0.01251220703125,
0.04766845703125,
-0.0231781005859375,
-0.049774169921875,
0.0596923828125,
0.0048828125,
0.07275390625,
-0.01873779296875,
0.006206512451171875,
0.00415802001953125,
0.0440673828125,
-0.0232086181640625,
-0.03759765625,
0.04132080078125,
-0.01033782958984375,
-0.005084991455078125,
-0.003841400146484375,
0.0347900390625,
-0.0302886962890625,
-0.033355712890625,
0.0027637481689453125,
0.02496337890625,
0.0274658203125,
0.0016183853149414062,
-0.058349609375,
0.0121917724609375,
0.00653839111328125,
-0.035369873046875,
-0.0010204315185546875,
0.00007396936416625977,
0.018402099609375,
0.05322265625,
0.056396484375,
0.0005283355712890625,
0.0194854736328125,
-0.019805908203125,
0.06622314453125,
-0.040863037109375,
-0.032867431640625,
-0.07977294921875,
0.0450439453125,
-0.0089263916015625,
-0.06646728515625,
0.0555419921875,
0.040618896484375,
0.053497314453125,
-0.0146484375,
0.05096435546875,
-0.0299072265625,
0.00910186767578125,
-0.0027923583984375,
0.0787353515625,
0.0110321044921875,
0.0008192062377929688,
-0.002811431884765625,
-0.061798095703125,
0.0141143798828125,
0.0531005859375,
-0.02423095703125,
0.0157928466796875,
0.048095703125,
0.055419921875,
0.0004725456237792969,
0.00010561943054199219,
0.007633209228515625,
0.0153350830078125,
0.041168212890625,
0.04388427734375,
0.037994384765625,
-0.06341552734375,
0.057769775390625,
-0.026519775390625,
-0.008056640625,
-0.00821685791015625,
-0.042572021484375,
-0.07574462890625,
-0.050872802734375,
-0.0169219970703125,
-0.043365478515625,
-0.0241241455078125,
0.08099365234375,
0.059814453125,
-0.06280517578125,
-0.0208587646484375,
-0.01371002197265625,
-0.0181884765625,
-0.004291534423828125,
-0.024658203125,
0.043060302734375,
-0.02276611328125,
-0.064453125,
-0.01023101806640625,
-0.0068817138671875,
0.0203857421875,
-0.031768798828125,
0.0009326934814453125,
-0.0148162841796875,
0.009979248046875,
0.009765625,
0.0123443603515625,
-0.061737060546875,
-0.0195159912109375,
-0.004741668701171875,
-0.022796630859375,
-0.0004417896270751953,
0.019134521484375,
-0.059326171875,
0.00884246826171875,
0.03924560546875,
0.0139617919921875,
0.07550048828125,
-0.024078369140625,
0.033782958984375,
-0.034210205078125,
0.0170135498046875,
0.01500701904296875,
0.0301513671875,
0.010589599609375,
-0.03961181640625,
0.00009238719940185547,
0.015167236328125,
-0.0498046875,
-0.046600341796875,
-0.005462646484375,
-0.06610107421875,
-0.01520538330078125,
0.06610107421875,
-0.02197265625,
-0.0266265869140625,
-0.010528564453125,
-0.03924560546875,
0.0701904296875,
-0.034088134765625,
0.06951904296875,
0.04266357421875,
-0.023773193359375,
-0.0174560546875,
-0.016326904296875,
0.0421142578125,
0.01507568359375,
-0.06842041015625,
-0.00201416015625,
0.0015888214111328125,
0.040924072265625,
0.018951416015625,
0.042236328125,
0.014739990234375,
0.0218658447265625,
0.011932373046875,
0.01285552978515625,
-0.031280517578125,
0.00281524658203125,
0.004608154296875,
0.005474090576171875,
-0.0231170654296875,
-0.045166015625
]
] |
microsoft/swinv2-tiny-patch4-window8-256 | 2022-12-10T10:01:54.000Z | [
"transformers",
"pytorch",
"swinv2",
"image-classification",
"vision",
"dataset:imagenet-1k",
"arxiv:2111.09883",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | image-classification | microsoft | null | null | microsoft/swinv2-tiny-patch4-window8-256 | 4 | 7,330 | transformers | 2022-06-14T06:00:27 | ---
license: apache-2.0
tags:
- vision
- image-classification
datasets:
- imagenet-1k
widget:
- src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/tiger.jpg
example_title: Tiger
- src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/teapot.jpg
example_title: Teapot
- src: https://huggingface.co/datasets/mishig/sample_images/resolve/main/palace.jpg
example_title: Palace
---
# Swin Transformer v2 (tiny-sized model)
Swin Transformer v2 model pre-trained on ImageNet-1k at resolution 256x256. It was introduced in the paper [Swin Transformer V2: Scaling Up Capacity and Resolution](https://arxiv.org/abs/2111.09883) by Liu et al. and first released in [this repository](https://github.com/microsoft/Swin-Transformer).
Disclaimer: The team releasing Swin Transformer v2 did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Swin Transformer is a type of Vision Transformer. It builds hierarchical feature maps by merging image patches (shown in gray) in deeper layers and has linear computation complexity to input image size due to computation of self-attention only within each local window (shown in red). It can thus serve as a general-purpose backbone for both image classification and dense recognition tasks. In contrast, previous vision Transformers produce feature maps of a single low resolution and have quadratic computation complexity to input image size due to computation of self-attention globally.
Swin Transformer v2 adds 3 main improvements: 1) a residual-post-norm method combined with cosine attention to improve training stability; 2) a log-spaced continuous position bias method to effectively transfer models pre-trained using low-resolution images to downstream tasks with high-resolution inputs; 3) a self-supervised pre-training method, SimMIM, to reduce the needs of vast labeled images.
[Source](https://paperswithcode.com/method/swin-transformer)
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=swinv2) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model to classify an image of the COCO 2017 dataset into one of the 1,000 ImageNet classes:
```python
from transformers import AutoImageProcessor, AutoModelForImageClassification
from PIL import Image
import requests
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
processor = AutoImageProcessor.from_pretrained("microsoft/swinv2-tiny-patch4-window8-256")
model = AutoModelForImageClassification.from_pretrained("microsoft/swinv2-tiny-patch4-window8-256")
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
logits = outputs.logits
# model predicts one of the 1000 ImageNet classes
predicted_class_idx = logits.argmax(-1).item()
print("Predicted class:", model.config.id2label[predicted_class_idx])
```
For more code examples, we refer to the [documentation](https://huggingface.co/transformers/model_doc/swinv2.html#).
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2111-09883,
author = {Ze Liu and
Han Hu and
Yutong Lin and
Zhuliang Yao and
Zhenda Xie and
Yixuan Wei and
Jia Ning and
Yue Cao and
Zheng Zhang and
Li Dong and
Furu Wei and
Baining Guo},
title = {Swin Transformer {V2:} Scaling Up Capacity and Resolution},
journal = {CoRR},
volume = {abs/2111.09883},
year = {2021},
url = {https://arxiv.org/abs/2111.09883},
eprinttype = {arXiv},
eprint = {2111.09883},
timestamp = {Thu, 02 Dec 2021 15:54:22 +0100},
biburl = {https://dblp.org/rec/journals/corr/abs-2111-09883.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
``` | 4,191 | [
[
-0.045623779296875,
-0.0181732177734375,
-0.0160980224609375,
0.01451873779296875,
-0.008636474609375,
-0.029296875,
-0.00437164306640625,
-0.063720703125,
0.005443572998046875,
0.0296478271484375,
-0.040069580078125,
0.00048804283142089844,
-0.0447998046875,
-0.01213836669921875,
-0.0258636474609375,
0.06170654296875,
-0.0008087158203125,
-0.004367828369140625,
-0.0335693359375,
-0.014678955078125,
-0.0254669189453125,
-0.0129547119140625,
-0.0439453125,
-0.0280303955078125,
0.03460693359375,
0.026611328125,
0.050140380859375,
0.035186767578125,
0.06298828125,
0.0335693359375,
-0.0010366439819335938,
-0.006740570068359375,
-0.0301055908203125,
-0.01153564453125,
0.00557708740234375,
-0.034912109375,
-0.03472900390625,
0.01259613037109375,
0.02392578125,
0.0185699462890625,
0.0197906494140625,
0.0379638671875,
0.00643157958984375,
0.0350341796875,
-0.0213165283203125,
0.01348876953125,
-0.0439453125,
0.0101470947265625,
-0.001148223876953125,
0.0015478134155273438,
-0.0275115966796875,
0.000583648681640625,
0.0174713134765625,
-0.0306549072265625,
0.05291748046875,
0.0177154541015625,
0.1029052734375,
0.01324462890625,
-0.023284912109375,
0.00989532470703125,
-0.044097900390625,
0.06976318359375,
-0.051666259765625,
0.01526641845703125,
0.01396942138671875,
0.034515380859375,
0.0137176513671875,
-0.07550048828125,
-0.035888671875,
0.0045928955078125,
-0.0306549072265625,
0.0172576904296875,
-0.0352783203125,
-0.0044097900390625,
0.0293121337890625,
0.0285491943359375,
-0.04010009765625,
0.026153564453125,
-0.0633544921875,
-0.0262908935546875,
0.06024169921875,
-0.0031223297119140625,
0.0181884765625,
-0.0232391357421875,
-0.05029296875,
-0.026641845703125,
-0.03277587890625,
0.0173492431640625,
0.0032863616943359375,
0.005397796630859375,
-0.029815673828125,
0.044586181640625,
0.01904296875,
0.0261993408203125,
0.018646240234375,
-0.01541900634765625,
0.03228759765625,
-0.0239715576171875,
-0.0118865966796875,
-0.00612640380859375,
0.06890869140625,
0.0447998046875,
0.0007905960083007812,
0.0211029052734375,
-0.01983642578125,
-0.012908935546875,
0.0241851806640625,
-0.08056640625,
-0.0170745849609375,
0.01125335693359375,
-0.04937744140625,
-0.036834716796875,
0.005199432373046875,
-0.043243408203125,
-0.00783538818359375,
-0.021484375,
0.0305023193359375,
-0.01013946533203125,
-0.035491943359375,
-0.035400390625,
-0.0204620361328125,
0.04461669921875,
0.0274505615234375,
-0.0506591796875,
0.01268768310546875,
0.0209197998046875,
0.07659912109375,
-0.01348114013671875,
-0.02166748046875,
0.0017042160034179688,
-0.0222625732421875,
-0.0216064453125,
0.051025390625,
0.0038433074951171875,
-0.01258087158203125,
-0.0091705322265625,
0.030029296875,
0.0026912689208984375,
-0.03729248046875,
0.0196685791015625,
-0.038299560546875,
0.01451873779296875,
-0.00878143310546875,
-0.0110931396484375,
-0.01070404052734375,
0.004093170166015625,
-0.050811767578125,
0.09539794921875,
0.0413818359375,
-0.07061767578125,
0.00800323486328125,
-0.03314208984375,
-0.027740478515625,
0.006389617919921875,
0.0117950439453125,
-0.05291748046875,
0.00322723388671875,
-0.0013380050659179688,
0.03338623046875,
-0.0243377685546875,
0.0148468017578125,
-0.0279998779296875,
-0.0211181640625,
0.005100250244140625,
-0.0213775634765625,
0.07000732421875,
0.0188751220703125,
-0.03289794921875,
0.01383209228515625,
-0.03515625,
0.00313568115234375,
0.039306640625,
0.0103759765625,
-0.0121917724609375,
-0.0268096923828125,
0.0257568359375,
0.04840087890625,
0.031341552734375,
-0.033416748046875,
0.0216217041015625,
-0.0234375,
0.0399169921875,
0.04547119140625,
-0.0133514404296875,
0.0472412109375,
-0.0160064697265625,
0.0254364013671875,
0.018035888671875,
0.0447998046875,
-0.0093231201171875,
-0.045745849609375,
-0.07611083984375,
-0.0033512115478515625,
0.0135955810546875,
0.035247802734375,
-0.047271728515625,
0.04119873046875,
-0.03118896484375,
-0.06036376953125,
-0.035369873046875,
0.00205230712890625,
0.01194000244140625,
0.038787841796875,
0.04400634765625,
-0.017822265625,
-0.06829833984375,
-0.08087158203125,
0.006591796875,
-0.004863739013671875,
-0.007289886474609375,
0.03033447265625,
0.056121826171875,
-0.0382080078125,
0.07147216796875,
-0.0168304443359375,
-0.0310211181640625,
-0.0311126708984375,
0.01031494140625,
0.01849365234375,
0.04071044921875,
0.048248291015625,
-0.061370849609375,
-0.035888671875,
-0.001800537109375,
-0.05718994140625,
0.005405426025390625,
-0.0102691650390625,
-0.0103912353515625,
0.023162841796875,
0.0222320556640625,
-0.03509521484375,
0.048004150390625,
0.0401611328125,
-0.0205078125,
0.04998779296875,
-0.007785797119140625,
0.0008983612060546875,
-0.0701904296875,
0.0016298294067382812,
0.0212554931640625,
-0.0101470947265625,
-0.04150390625,
-0.00998687744140625,
0.02197265625,
-0.00261688232421875,
-0.048248291015625,
0.046295166015625,
-0.035186767578125,
-0.01302337646484375,
-0.0222930908203125,
0.0027408599853515625,
0.00579071044921875,
0.042236328125,
0.01224517822265625,
0.03485107421875,
0.0477294921875,
-0.0347900390625,
0.0321044921875,
0.0240478515625,
-0.023681640625,
0.023468017578125,
-0.07513427734375,
-0.00091552734375,
-0.0001304149627685547,
0.0261993408203125,
-0.07281494140625,
-0.007526397705078125,
-0.00016176700592041016,
-0.03753662109375,
0.043182373046875,
-0.026275634765625,
-0.020782470703125,
-0.07452392578125,
-0.0287322998046875,
0.03369140625,
0.05828857421875,
-0.0543212890625,
0.040924072265625,
0.01317596435546875,
0.004314422607421875,
-0.044921875,
-0.08648681640625,
-0.01496124267578125,
0.002803802490234375,
-0.06561279296875,
0.0343017578125,
0.00444793701171875,
-0.0003578662872314453,
0.0073089599609375,
-0.017791748046875,
0.00307464599609375,
-0.0182647705078125,
0.03765869140625,
0.057830810546875,
-0.0175933837890625,
-0.01500701904296875,
0.007770538330078125,
-0.017669677734375,
0.00849151611328125,
-0.00763702392578125,
0.029815673828125,
-0.03387451171875,
-0.0059814453125,
-0.035614013671875,
0.002254486083984375,
0.05029296875,
-0.00974273681640625,
0.042724609375,
0.0809326171875,
-0.020721435546875,
-0.0018281936645507812,
-0.045166015625,
-0.0280609130859375,
-0.041778564453125,
0.0278472900390625,
-0.0203094482421875,
-0.05157470703125,
0.0401611328125,
0.002819061279296875,
0.0139617919921875,
0.07122802734375,
0.0241851806640625,
-0.0257415771484375,
0.07391357421875,
0.0282440185546875,
0.00931549072265625,
0.04486083984375,
-0.0660400390625,
0.0172119140625,
-0.07257080078125,
-0.03619384765625,
-0.027618408203125,
-0.04986572265625,
-0.04754638671875,
-0.046630859375,
0.0197296142578125,
0.01363372802734375,
-0.030303955078125,
0.059906005859375,
-0.05633544921875,
0.00798797607421875,
0.048492431640625,
0.005847930908203125,
-0.02239990234375,
0.016357421875,
0.004573822021484375,
-0.0114593505859375,
-0.05499267578125,
-0.0034313201904296875,
0.056182861328125,
0.043121337890625,
0.050018310546875,
-0.00797271728515625,
0.03338623046875,
0.0216217041015625,
0.01824951171875,
-0.0615234375,
0.039764404296875,
-0.0027408599853515625,
-0.0496826171875,
-0.01544952392578125,
-0.0167083740234375,
-0.07220458984375,
0.0220794677734375,
-0.023681640625,
-0.043548583984375,
0.040863037109375,
0.0120086669921875,
0.00818634033203125,
0.037200927734375,
-0.05230712890625,
0.06646728515625,
-0.029876708984375,
-0.033294677734375,
-0.00463104248046875,
-0.07037353515625,
0.0218505859375,
0.01751708984375,
0.00432586669921875,
-0.0017642974853515625,
0.01218414306640625,
0.0643310546875,
-0.050445556640625,
0.07281494140625,
-0.0250091552734375,
0.017486572265625,
0.047698974609375,
-0.005878448486328125,
0.050537109375,
-0.01204681396484375,
0.0121612548828125,
0.0457763671875,
0.0029010772705078125,
-0.0347900390625,
-0.044097900390625,
0.0535888671875,
-0.080078125,
-0.0325927734375,
-0.035064697265625,
-0.019134521484375,
0.01462554931640625,
0.023529052734375,
0.0567626953125,
0.03759765625,
0.009063720703125,
0.03155517578125,
0.0447998046875,
-0.007568359375,
0.042816162109375,
0.0117340087890625,
-0.015289306640625,
-0.017303466796875,
0.06146240234375,
0.0211029052734375,
0.0147705078125,
0.0182342529296875,
0.0235443115234375,
-0.0299224853515625,
-0.018463134765625,
-0.022796630859375,
0.0231475830078125,
-0.05084228515625,
-0.048431396484375,
-0.04833984375,
-0.04638671875,
-0.050384521484375,
-0.02093505859375,
-0.042327880859375,
-0.024200439453125,
-0.032470703125,
0.0027561187744140625,
0.0308990478515625,
0.048065185546875,
-0.003307342529296875,
0.0144195556640625,
-0.0362548828125,
0.01160430908203125,
0.0250701904296875,
0.0240478515625,
0.007480621337890625,
-0.0721435546875,
-0.01340484619140625,
0.0009131431579589844,
-0.0285491943359375,
-0.04217529296875,
0.036163330078125,
0.00965118408203125,
0.038177490234375,
0.03826904296875,
0.00567626953125,
0.051910400390625,
-0.0215911865234375,
0.06732177734375,
0.05548095703125,
-0.048187255859375,
0.05340576171875,
0.00176239013671875,
0.0216217041015625,
0.0104827880859375,
0.0282745361328125,
-0.0285491943359375,
-0.00594329833984375,
-0.069580078125,
-0.072021484375,
0.0579833984375,
0.00560760498046875,
0.00350189208984375,
0.01763916015625,
0.0161285400390625,
0.0012845993041992188,
-0.00909423828125,
-0.05731201171875,
-0.04437255859375,
-0.05377197265625,
-0.006927490234375,
-0.006114959716796875,
-0.035125732421875,
-0.0018281936645507812,
-0.059234619140625,
0.05352783203125,
-0.00708770751953125,
0.05963134765625,
0.02020263671875,
-0.0229949951171875,
-0.009368896484375,
-0.01029205322265625,
0.032562255859375,
0.015716552734375,
-0.00589752197265625,
0.00878143310546875,
0.01751708984375,
-0.04833984375,
-0.00679779052734375,
-0.004024505615234375,
-0.01448822021484375,
-0.0014829635620117188,
0.039764404296875,
0.0860595703125,
0.022857666015625,
-0.01529693603515625,
0.07061767578125,
0.0013179779052734375,
-0.044036865234375,
-0.04193115234375,
0.00933074951171875,
-0.006214141845703125,
0.0251617431640625,
0.0298614501953125,
0.048980712890625,
0.00783538818359375,
-0.017852783203125,
0.00501251220703125,
0.01525115966796875,
-0.039154052734375,
-0.0261688232421875,
0.048858642578125,
0.006114959716796875,
-0.01499176025390625,
0.0653076171875,
-0.00435638427734375,
-0.039581298828125,
0.06756591796875,
0.05914306640625,
0.046905517578125,
-0.006389617919921875,
0.0114593505859375,
0.0653076171875,
0.03118896484375,
-0.01088714599609375,
-0.006885528564453125,
0.005252838134765625,
-0.051483154296875,
-0.006816864013671875,
-0.03594970703125,
-0.0009927749633789062,
0.0166473388671875,
-0.06121826171875,
0.041168212890625,
-0.0230712890625,
-0.0207061767578125,
-0.0037841796875,
0.02191162109375,
-0.0833740234375,
0.0187835693359375,
0.01248931884765625,
0.0789794921875,
-0.05865478515625,
0.0650634765625,
0.043182373046875,
-0.0301971435546875,
-0.07073974609375,
-0.046875,
-0.003643035888671875,
-0.0611572265625,
0.03533935546875,
0.035491943359375,
0.0014905929565429688,
-0.0044097900390625,
-0.07037353515625,
-0.0709228515625,
0.1263427734375,
-0.0067138671875,
-0.050262451171875,
0.0042266845703125,
-0.00911712646484375,
0.03863525390625,
-0.0251312255859375,
0.048553466796875,
0.02386474609375,
0.046905517578125,
0.0285186767578125,
-0.04156494140625,
0.0115814208984375,
-0.04254150390625,
0.0269775390625,
-0.00197601318359375,
-0.05865478515625,
0.051727294921875,
-0.0418701171875,
-0.00759124755859375,
0.000881195068359375,
0.06903076171875,
-0.0030956268310546875,
0.01812744140625,
0.045623779296875,
0.0281829833984375,
0.041107177734375,
-0.030181884765625,
0.0726318359375,
-0.0047149658203125,
0.044708251953125,
0.07000732421875,
0.01296234130859375,
0.052978515625,
0.0302581787109375,
-0.02825927734375,
0.04693603515625,
0.0545654296875,
-0.054168701171875,
0.0247344970703125,
-0.004497528076171875,
0.0185394287109375,
-0.003467559814453125,
0.00501251220703125,
-0.04339599609375,
0.0207061767578125,
0.0220489501953125,
-0.034210205078125,
0.01213836669921875,
0.01934814453125,
-0.0178070068359375,
-0.0361328125,
-0.0234375,
0.036712646484375,
-0.0005469322204589844,
-0.036407470703125,
0.048004150390625,
-0.01006317138671875,
0.07183837890625,
-0.03948974609375,
0.01436614990234375,
-0.0193939208984375,
0.005069732666015625,
-0.03253173828125,
-0.056060791015625,
0.0146484375,
-0.0182647705078125,
-0.015838623046875,
-0.007389068603515625,
0.08953857421875,
-0.023956298828125,
-0.039093017578125,
0.0279541015625,
0.003246307373046875,
0.02197265625,
0.011474609375,
-0.07391357421875,
0.011322021484375,
-0.0030765533447265625,
-0.0462646484375,
0.02777099609375,
0.00628662109375,
-0.0016088485717773438,
0.052978515625,
0.038177490234375,
-0.01328277587890625,
0.0096588134765625,
0.002162933349609375,
0.05877685546875,
-0.038665771484375,
-0.0169830322265625,
-0.0209197998046875,
0.049224853515625,
-0.023406982421875,
-0.03009033203125,
0.052459716796875,
0.03826904296875,
0.058258056640625,
-0.0203857421875,
0.056671142578125,
-0.034912109375,
0.005092620849609375,
0.0090789794921875,
0.046722412109375,
-0.055908203125,
-0.00760650634765625,
-0.00443267822265625,
-0.052886962890625,
-0.0139617919921875,
0.05078125,
-0.0216522216796875,
0.01284027099609375,
0.04266357421875,
0.06134033203125,
-0.00750732421875,
-0.005878448486328125,
0.023040771484375,
0.023040771484375,
0.005619049072265625,
0.01284027099609375,
0.035614013671875,
-0.0726318359375,
0.041839599609375,
-0.057281494140625,
-0.01505279541015625,
-0.0443115234375,
-0.041900634765625,
-0.06475830078125,
-0.05828857421875,
-0.0309906005859375,
-0.053924560546875,
-0.034454345703125,
0.055389404296875,
0.0740966796875,
-0.0701904296875,
0.004238128662109375,
-0.0030956268310546875,
-0.00045990943908691406,
-0.037628173828125,
-0.022430419921875,
0.027191162109375,
-0.0100555419921875,
-0.04632568359375,
-0.0006513595581054688,
0.0024242401123046875,
0.02630615234375,
-0.0247802734375,
-0.0278472900390625,
0.00009316205978393555,
-0.00565338134765625,
0.054595947265625,
0.0280914306640625,
-0.049957275390625,
-0.0177459716796875,
0.00604248046875,
-0.025177001953125,
0.01165008544921875,
0.050994873046875,
-0.038055419921875,
0.0196990966796875,
0.047027587890625,
0.01444244384765625,
0.065673828125,
-0.01416778564453125,
-0.0013647079467773438,
-0.043701171875,
0.025970458984375,
0.01160430908203125,
0.03656005859375,
0.0159454345703125,
-0.026123046875,
0.03857421875,
0.0298004150390625,
-0.04571533203125,
-0.05523681640625,
0.0006780624389648438,
-0.103271484375,
-0.019287109375,
0.086669921875,
-0.00952911376953125,
-0.0309295654296875,
0.00766754150390625,
-0.011474609375,
0.02703857421875,
-0.01096343994140625,
0.02655029296875,
0.0162506103515625,
0.00217437744140625,
-0.037506103515625,
-0.031707763671875,
0.0189666748046875,
0.0003409385681152344,
-0.0345458984375,
-0.02362060546875,
0.0081939697265625,
0.032501220703125,
0.0236968994140625,
0.01494598388671875,
-0.0216217041015625,
0.0197906494140625,
0.021575927734375,
0.0369873046875,
-0.01611328125,
-0.0269317626953125,
-0.01366424560546875,
-0.005542755126953125,
-0.019134521484375,
-0.001598358154296875
]
] |
ehartford/samantha-mistral-instruct-7b | 2023-09-30T16:30:14.000Z | [
"transformers",
"pytorch",
"mistral",
"text-generation",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | ehartford | null | null | ehartford/samantha-mistral-instruct-7b | 18 | 7,324 | transformers | 2023-09-30T06:12:21 | ---
license: apache-2.0
---
Trained on [mistral-7b-instruct](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) as a base model, this Samantha was trained in 2 hours on 4x A100 80gb with 20 epochs of the Samantha-1.1 dataset.
My gratitude goes to my sponsors, a16z and Microsoft for Startups.
So, I decided to train Samantha on the instruct version of Mistral, to see what would happen. It turns out she's a very good and capable general assistant. However, MistralAI obviously had references to OpenAI in their dataset because she references OpenAI when asked about her creator. (so I trained another Samantha on mistral-base too)
NEW! This Samantha is trained using ChatML prompt format instead of Vicuna-1.1. All of my future models will use ChatML prompt format.
[Meet Samantha](https://erichartford.com/meet-samantha)
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.
Her conversation format is ChatML
https://github.com/openai/openai-python/blob/main/chatml.md
Prompt Template:
```
<|im_start|>system
You are a caring and empathetic sentient AI companion named Samantha.<|im_end|>
<|im_start|>user
Hello, what is your name?<|im_end|>
```
Example:
Official character card: (thanks MortalWombat)
Shout out and much thanks to WingLian, author of axolotl! And everyone who has contributed to the project.
[<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)
And much thanks as always to TheBloke for distribution.
| 2,186 | [
[
-0.0234375,
-0.0252685546875,
0.033782958984375,
-0.0018320083618164062,
-0.037811279296875,
-0.03631591796875,
0.0032958984375,
-0.035919189453125,
0.028594970703125,
0.0290985107421875,
-0.050140380859375,
-0.0290679931640625,
-0.0297393798828125,
0.009368896484375,
0.0172882080078125,
0.071533203125,
0.005153656005859375,
0.0160675048828125,
-0.007526397705078125,
-0.033782958984375,
-0.051055908203125,
-0.05078125,
-0.0787353515625,
-0.03387451171875,
0.041839599609375,
0.006778717041015625,
0.06439208984375,
0.038848876953125,
0.0186004638671875,
0.021575927734375,
-0.006954193115234375,
0.0283355712890625,
-0.024322509765625,
-0.0012617111206054688,
-0.0022430419921875,
-0.04046630859375,
-0.049896240234375,
0.0188751220703125,
0.0218505859375,
0.02001953125,
-0.010009765625,
0.01409149169921875,
-0.01059722900390625,
0.035400390625,
-0.041961669921875,
0.0177764892578125,
-0.018310546875,
0.00489044189453125,
0.001766204833984375,
0.00830841064453125,
-0.0291595458984375,
-0.0297393798828125,
-0.006481170654296875,
-0.0628662109375,
0.006183624267578125,
0.004852294921875,
0.0736083984375,
0.034576416015625,
-0.0225067138671875,
-0.00255584716796875,
-0.0574951171875,
0.046966552734375,
-0.039306640625,
0.0084686279296875,
0.031951904296875,
0.055999755859375,
-0.01287078857421875,
-0.054656982421875,
-0.049285888671875,
-0.0148468017578125,
-0.0025196075439453125,
0.00930023193359375,
-0.041656494140625,
0.004474639892578125,
0.0214080810546875,
0.0188140869140625,
-0.04388427734375,
-0.008575439453125,
-0.016937255859375,
-0.006610870361328125,
0.031341552734375,
0.0256500244140625,
0.036956787109375,
0.005252838134765625,
-0.02752685546875,
-0.018646240234375,
-0.034698486328125,
0.0188140869140625,
0.015380859375,
0.0127105712890625,
-0.038360595703125,
0.029998779296875,
-0.00768280029296875,
0.042724609375,
0.02032470703125,
0.005649566650390625,
0.0291900634765625,
-0.01338958740234375,
-0.023651123046875,
-0.0030536651611328125,
0.06494140625,
0.03302001953125,
0.034149169921875,
-0.010223388671875,
0.002574920654296875,
0.0185699462890625,
0.0212554931640625,
-0.0692138671875,
-0.0271148681640625,
0.04669189453125,
-0.055633544921875,
-0.01386260986328125,
0.01061248779296875,
-0.02490234375,
-0.04669189453125,
-0.025909423828125,
0.0251922607421875,
-0.057342529296875,
-0.03704833984375,
-0.0139312744140625,
-0.027679443359375,
0.0181121826171875,
0.0538330078125,
-0.059326171875,
0.004150390625,
0.038330078125,
0.0667724609375,
0.0137481689453125,
-0.029571533203125,
-0.037384033203125,
-0.0007271766662597656,
-0.0221710205078125,
0.048553466796875,
-0.050140380859375,
-0.025115966796875,
-0.0230712890625,
0.0107879638671875,
0.00412750244140625,
-0.060211181640625,
0.030517578125,
-0.0107421875,
0.01459503173828125,
-0.0239715576171875,
-0.0103759765625,
-0.0023059844970703125,
0.00214385986328125,
-0.03717041015625,
0.05194091796875,
0.027801513671875,
-0.045379638671875,
0.023468017578125,
-0.05438232421875,
-0.0010900497436523438,
-0.005702972412109375,
-0.00939178466796875,
-0.0091705322265625,
-0.00519561767578125,
0.019927978515625,
0.02899169921875,
-0.0296783447265625,
0.0011653900146484375,
-0.030853271484375,
-0.027587890625,
0.03570556640625,
-0.0169219970703125,
0.0814208984375,
0.0232696533203125,
-0.01313018798828125,
0.019744873046875,
-0.067626953125,
0.00479888916015625,
0.01363372802734375,
0.003749847412109375,
-0.030853271484375,
-0.0283966064453125,
-0.01253509521484375,
0.0256500244140625,
0.01151275634765625,
-0.0418701171875,
0.0247802734375,
-0.0168609619140625,
0.0252685546875,
0.06292724609375,
0.004299163818359375,
0.041168212890625,
-0.04534912109375,
0.04296875,
0.01277923583984375,
0.04949951171875,
-0.0295867919921875,
-0.042694091796875,
-0.050323486328125,
-0.04681396484375,
-0.0000795125961303711,
0.0250701904296875,
-0.040313720703125,
0.03155517578125,
0.005123138427734375,
-0.067626953125,
-0.0723876953125,
-0.02105712890625,
0.03228759765625,
0.0227203369140625,
0.0350341796875,
-0.032379150390625,
-0.037139892578125,
-0.0489501953125,
0.01056671142578125,
-0.039886474609375,
-0.0130157470703125,
0.027496337890625,
0.0279693603515625,
-0.0294036865234375,
0.0806884765625,
-0.03997802734375,
-0.01438140869140625,
-0.0036487579345703125,
-0.01519775390625,
0.0252685546875,
0.0560302734375,
0.058197021484375,
-0.042388916015625,
-0.02490234375,
0.00936126708984375,
-0.08233642578125,
0.004825592041015625,
-0.01019287109375,
-0.05767822265625,
0.00843048095703125,
0.0070648193359375,
-0.0926513671875,
0.04705810546875,
0.007293701171875,
-0.0439453125,
0.044036865234375,
-0.01558685302734375,
0.013763427734375,
-0.10296630859375,
0.0239715576171875,
-0.0193023681640625,
-0.006252288818359375,
-0.0498046875,
0.01666259765625,
-0.01551055908203125,
-0.024139404296875,
-0.026397705078125,
0.07171630859375,
-0.038848876953125,
0.0088043212890625,
-0.0184783935546875,
-0.01517486572265625,
-0.0145721435546875,
0.05950927734375,
-0.006145477294921875,
0.0496826171875,
0.05133056640625,
-0.044189453125,
0.047821044921875,
0.05047607421875,
0.0151214599609375,
0.07659912109375,
-0.066650390625,
0.02734375,
-0.0264739990234375,
0.03118896484375,
-0.07379150390625,
-0.0197296142578125,
0.06829833984375,
-0.06121826171875,
0.0105743408203125,
-0.015655517578125,
-0.0262603759765625,
-0.01116943359375,
-0.0121612548828125,
0.01233673095703125,
0.060302734375,
-0.06341552734375,
0.0565185546875,
0.01290130615234375,
0.0033664703369140625,
-0.0292510986328125,
-0.039642333984375,
-0.00392913818359375,
-0.0177764892578125,
-0.052490234375,
0.0095977783203125,
-0.0177001953125,
-0.0309906005859375,
-0.003833770751953125,
-0.013092041015625,
-0.02490234375,
-0.0102081298828125,
0.06341552734375,
0.0287322998046875,
-0.0168609619140625,
0.002819061279296875,
-0.007808685302734375,
-0.01629638671875,
0.0011844635009765625,
-0.0141143798828125,
0.058837890625,
-0.01003265380859375,
-0.0258941650390625,
-0.062042236328125,
0.017181396484375,
0.05072021484375,
-0.0261993408203125,
0.0736083984375,
0.0537109375,
-0.024078369140625,
0.0033473968505859375,
-0.0252685546875,
-0.00899505615234375,
-0.03369140625,
-0.00722503662109375,
-0.0169677734375,
-0.0340576171875,
0.05511474609375,
0.016510009765625,
0.01288604736328125,
0.0289459228515625,
0.0304718017578125,
0.0026683807373046875,
0.07489013671875,
0.0469970703125,
-0.0223236083984375,
0.04400634765625,
-0.0179443359375,
-0.009796142578125,
-0.05987548828125,
-0.039337158203125,
-0.04095458984375,
-0.017486572265625,
-0.03125,
-0.0251922607421875,
0.02215576171875,
-0.0014791488647460938,
-0.051727294921875,
0.032440185546875,
-0.04034423828125,
0.0123291015625,
0.0379638671875,
0.03668212890625,
0.006103515625,
-0.00745391845703125,
0.018402099609375,
0.01517486572265625,
-0.037933349609375,
-0.04156494140625,
0.0670166015625,
0.0322265625,
0.06390380859375,
0.0303192138671875,
0.05999755859375,
0.01611328125,
0.0117645263671875,
-0.0295867919921875,
0.045135498046875,
0.01904296875,
-0.04266357421875,
-0.017913818359375,
-0.0257415771484375,
-0.08673095703125,
0.0097198486328125,
0.0015916824340820312,
-0.05615234375,
0.023590087890625,
0.0113067626953125,
-0.029937744140625,
0.00031256675720214844,
-0.07452392578125,
0.060516357421875,
0.0108642578125,
-0.0132598876953125,
-0.004306793212890625,
-0.059967041015625,
0.0140838623046875,
0.020263671875,
-0.0216827392578125,
0.004154205322265625,
-0.0033817291259765625,
0.034393310546875,
-0.08111572265625,
0.07958984375,
-0.03521728515625,
0.0092620849609375,
0.030975341796875,
-0.009765625,
0.0258636474609375,
0.00820159912109375,
-0.007236480712890625,
-0.000820159912109375,
0.007579803466796875,
-0.04534912109375,
-0.05615234375,
0.0280609130859375,
-0.0968017578125,
-0.0227508544921875,
-0.037933349609375,
-0.004241943359375,
0.005039215087890625,
0.00276947021484375,
0.0278472900390625,
0.039520263671875,
-0.022674560546875,
-0.014068603515625,
0.03753662109375,
-0.026092529296875,
0.024017333984375,
0.019256591796875,
-0.00844573974609375,
-0.0377197265625,
0.0699462890625,
-0.01776123046875,
0.0153656005859375,
0.0132598876953125,
0.0025959014892578125,
-0.002300262451171875,
0.00428009033203125,
-0.04827880859375,
0.0297393798828125,
-0.054046630859375,
-0.013458251953125,
-0.046295166015625,
-0.0220947265625,
-0.0382080078125,
-0.006618499755859375,
-0.0130767822265625,
-0.0260772705078125,
-0.04840087890625,
0.0185699462890625,
0.044036865234375,
0.0596923828125,
0.0161590576171875,
0.036865234375,
-0.050628662109375,
0.009429931640625,
0.0232391357421875,
0.01329803466796875,
0.0218963623046875,
-0.04595947265625,
-0.004497528076171875,
0.006305694580078125,
-0.0164031982421875,
-0.062042236328125,
0.0207366943359375,
0.002857208251953125,
0.06488037109375,
0.0386962890625,
-0.001430511474609375,
0.04034423828125,
-0.020843505859375,
0.07037353515625,
0.01554107666015625,
-0.03436279296875,
0.03680419921875,
-0.039703369140625,
0.03564453125,
0.0307769775390625,
0.03680419921875,
-0.04046630859375,
-0.0186309814453125,
-0.057159423828125,
-0.0291900634765625,
0.06353759765625,
0.026824951171875,
0.0174713134765625,
0.00662994384765625,
0.035614013671875,
0.015380859375,
0.0252532958984375,
-0.03839111328125,
-0.034820556640625,
-0.0302581787109375,
-0.00897979736328125,
-0.0039215087890625,
-0.00835418701171875,
0.0019407272338867188,
-0.0252227783203125,
0.05694580078125,
-0.0092010498046875,
0.06829833984375,
0.023651123046875,
0.0120697021484375,
-0.00424957275390625,
-0.0092620849609375,
0.03521728515625,
0.0298919677734375,
-0.0196990966796875,
-0.01128387451171875,
-0.0007500648498535156,
-0.046600341796875,
0.0089874267578125,
0.0078582763671875,
0.003208160400390625,
0.01490020751953125,
0.036376953125,
0.08758544921875,
-0.0328369140625,
-0.040802001953125,
0.041351318359375,
-0.0243377685546875,
0.012969970703125,
-0.0357666015625,
0.0251922607421875,
-0.007442474365234375,
0.039703369140625,
0.01290130615234375,
0.03509521484375,
-0.00856781005859375,
-0.053436279296875,
-0.0031757354736328125,
0.022918701171875,
-0.03436279296875,
-0.0474853515625,
0.057159423828125,
0.006999969482421875,
-0.0288848876953125,
0.040802001953125,
-0.014373779296875,
-0.024505615234375,
0.04876708984375,
0.04071044921875,
0.07568359375,
-0.038543701171875,
0.02935791015625,
0.038360595703125,
0.01038360595703125,
0.01412200927734375,
0.036865234375,
-0.01058197021484375,
-0.026275634765625,
0.0153045654296875,
-0.039886474609375,
-0.041778564453125,
0.0012798309326171875,
-0.0265655517578125,
0.033477783203125,
-0.06939697265625,
-0.01233673095703125,
0.003513336181640625,
-0.01384735107421875,
-0.05316162109375,
0.02069091796875,
-0.00844573974609375,
0.07769775390625,
-0.052978515625,
0.045318603515625,
0.0701904296875,
-0.05731201171875,
-0.07354736328125,
0.01128387451171875,
-0.0015954971313476562,
-0.059600830078125,
0.030181884765625,
0.00939178466796875,
0.015899658203125,
-0.00896453857421875,
-0.058013916015625,
-0.04876708984375,
0.09722900390625,
0.024993896484375,
0.0044708251953125,
-0.0108642578125,
-0.007617950439453125,
0.054290771484375,
-0.033966064453125,
0.06414794921875,
0.0143280029296875,
0.0236968994140625,
0.01232147216796875,
-0.06842041015625,
0.01490020751953125,
-0.05078125,
-0.0033054351806640625,
-0.0024776458740234375,
-0.082763671875,
0.08489990234375,
-0.0112762451171875,
-0.005840301513671875,
0.0489501953125,
0.06402587890625,
0.00946044921875,
0.0284423828125,
0.033294677734375,
0.023529052734375,
0.0706787109375,
0.006320953369140625,
0.08087158203125,
-0.01007080078125,
-0.0137939453125,
0.06341552734375,
-0.00418853759765625,
0.0439453125,
0.01361083984375,
-0.013427734375,
0.037109375,
0.056549072265625,
0.0019779205322265625,
0.0262451171875,
-0.011077880859375,
-0.026153564453125,
-0.006313323974609375,
-0.03656005859375,
-0.038848876953125,
0.0222320556640625,
-0.0113372802734375,
-0.031646728515625,
-0.0072479248046875,
0.01309967041015625,
0.002655029296875,
-0.004329681396484375,
-0.01873779296875,
0.048980712890625,
0.0033206939697265625,
-0.055450439453125,
0.06390380859375,
-0.01036834716796875,
0.041900634765625,
-0.053863525390625,
-0.0166015625,
-0.0310211181640625,
0.01294708251953125,
-0.0158233642578125,
-0.035430908203125,
-0.00754547119140625,
0.001800537109375,
-0.0058135986328125,
0.0011606216430664062,
0.05047607421875,
-0.0146331787109375,
-0.004878997802734375,
-0.0027828216552734375,
0.032135009765625,
0.047119140625,
-0.01486968994140625,
-0.032562255859375,
0.004711151123046875,
0.0082550048828125,
0.0113372802734375,
0.0301055908203125,
0.04595947265625,
-0.01336669921875,
0.045379638671875,
0.04345703125,
-0.0113372802734375,
-0.020782470703125,
-0.0015687942504882812,
0.0831298828125,
-0.03900146484375,
-0.0421142578125,
-0.060394287109375,
0.044342041015625,
-0.0015249252319335938,
-0.06329345703125,
0.046356201171875,
0.0310211181640625,
0.036285400390625,
-0.0246734619140625,
0.056976318359375,
-0.0162353515625,
0.025177001953125,
-0.036224365234375,
0.0574951171875,
-0.04669189453125,
-0.01108551025390625,
-0.032440185546875,
-0.06317138671875,
0.016510009765625,
0.041107177734375,
0.0076904296875,
0.0233917236328125,
0.031768798828125,
0.0626220703125,
-0.0051727294921875,
0.00563812255859375,
0.0169830322265625,
-0.0030117034912109375,
0.0206451416015625,
0.04486083984375,
0.070068359375,
-0.0234832763671875,
0.03643798828125,
-0.004268646240234375,
-0.03411865234375,
-0.01190948486328125,
-0.0193328857421875,
-0.11444091796875,
-0.058135986328125,
-0.0037746429443359375,
-0.0382080078125,
0.0111236572265625,
0.1005859375,
0.06134033203125,
-0.03802490234375,
-0.004665374755859375,
0.004795074462890625,
-0.01221466064453125,
-0.01398468017578125,
-0.01140594482421875,
0.01117706298828125,
0.00510406494140625,
-0.052154541015625,
0.0166168212890625,
-0.0034046173095703125,
0.033416748046875,
-0.01611328125,
-0.0263671875,
0.0004978179931640625,
-0.0030345916748046875,
0.0223541259765625,
0.03289794921875,
-0.044647216796875,
-0.0248870849609375,
0.0158233642578125,
-0.02655029296875,
0.01290130615234375,
0.0229034423828125,
-0.06011962890625,
0.0177154541015625,
0.0250701904296875,
0.03619384765625,
0.0218963623046875,
0.0212249755859375,
0.04571533203125,
-0.0408935546875,
0.025360107421875,
0.0033893585205078125,
0.015045166015625,
0.0209503173828125,
-0.046051025390625,
0.05828857421875,
0.027862548828125,
-0.044891357421875,
-0.056060791015625,
0.01207733154296875,
-0.08172607421875,
-0.01399993896484375,
0.08978271484375,
-0.0134429931640625,
-0.0350341796875,
0.01259613037109375,
-0.07525634765625,
0.028564453125,
-0.0518798828125,
0.0257415771484375,
0.03948974609375,
-0.0139617919921875,
-0.00626373291015625,
-0.02508544921875,
0.03436279296875,
0.0262603759765625,
-0.058807373046875,
-0.002532958984375,
0.040313720703125,
0.005985260009765625,
0.0261993408203125,
0.0762939453125,
0.0135955810546875,
0.040130615234375,
0.021087646484375,
0.0258636474609375,
-0.0280303955078125,
-0.0224151611328125,
-0.038970947265625,
-0.0244140625,
0.01480865478515625,
-0.048370361328125
]
] |
facebook/dino-vits8 | 2023-05-22T07:04:27.000Z | [
"transformers",
"pytorch",
"vit",
"feature-extraction",
"dino",
"vision",
"dataset:imagenet-1k",
"arxiv:2104.14294",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | feature-extraction | facebook | null | null | facebook/dino-vits8 | 9 | 7,318 | transformers | 2022-03-02T23:29:05 | ---
license: apache-2.0
tags:
- dino
- vision
datasets:
- imagenet-1k
---
# Vision Transformer (small-sized model, patch size 8) trained using DINO
Vision Transformer (ViT) model trained using the DINO method. It was introduced in the paper [Emerging Properties in Self-Supervised Vision Transformers](https://arxiv.org/abs/2104.14294) by Mathilde Caron, Hugo Touvron, Ishan Misra, Hervé Jégou, Julien Mairal, Piotr Bojanowski, Armand Joulin and first released in [this repository](https://github.com/facebookresearch/dino).
Disclaimer: The team releasing DINO did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a self-supervised fashion, namely ImageNet-1k, at a resolution of 224x224 pixels.
Images are presented to the model as a sequence of fixed-size patches (resolution 8x8), which are linearly embedded. One also adds a [CLS] token to the beginning of a sequence to use it for classification tasks. One also adds absolute position embeddings before feeding the sequence to the layers of the Transformer encoder.
Note that this model does not include any fine-tuned heads.
By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
## Intended uses & limitations
You can use the raw model for image classification. See the [model hub](https://huggingface.co/models?search=google/vit) to look for
fine-tuned versions on a task that interests you.
### How to use
Here is how to use this model:
```python
from transformers import ViTImageProcessor, ViTModel
from PIL import Image
import requests
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = ViTImageProcessor.from_pretrained('facebook/dino-vits8')
model = ViTModel.from_pretrained('facebook/dino-vits8')
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
last_hidden_states = outputs.last_hidden_state
```
### BibTeX entry and citation info
```bibtex
@article{DBLP:journals/corr/abs-2104-14294,
author = {Mathilde Caron and
Hugo Touvron and
Ishan Misra and
Herv{\'{e}} J{\'{e}}gou and
Julien Mairal and
Piotr Bojanowski and
Armand Joulin},
title = {Emerging Properties in Self-Supervised Vision Transformers},
journal = {CoRR},
volume = {abs/2104.14294},
year = {2021},
url = {https://arxiv.org/abs/2104.14294},
archivePrefix = {arXiv},
eprint = {2104.14294},
timestamp = {Tue, 04 May 2021 15:12:43 +0200},
biburl = {https://dblp.org/rec/journals/corr/abs-2104-14294.bib},
bibsource = {dblp computer science bibliography, https://dblp.org}
}
``` | 3,255 | [
[
-0.037994384765625,
-0.0182037353515625,
0.00946807861328125,
-0.0078125,
-0.0301971435546875,
-0.0018463134765625,
0.00681304931640625,
-0.038482666015625,
0.0256195068359375,
0.036224365234375,
-0.0313720703125,
-0.016021728515625,
-0.045196533203125,
-0.007755279541015625,
-0.036407470703125,
0.07159423828125,
-0.0030460357666015625,
-0.00926971435546875,
-0.01389312744140625,
-0.005954742431640625,
-0.015960693359375,
-0.0494384765625,
-0.041290283203125,
-0.0306549072265625,
0.033172607421875,
-0.002216339111328125,
0.05145263671875,
0.06591796875,
0.038726806640625,
0.034881591796875,
-0.0101776123046875,
-0.00201416015625,
-0.03326416015625,
-0.0208282470703125,
-0.015594482421875,
-0.03466796875,
-0.02276611328125,
0.01348114013671875,
0.046112060546875,
0.02996826171875,
0.017333984375,
0.021331787109375,
0.0032176971435546875,
0.015350341796875,
-0.04437255859375,
0.0271148681640625,
-0.034881591796875,
0.03228759765625,
-0.005863189697265625,
-0.005313873291015625,
-0.02911376953125,
-0.02276611328125,
0.0159912109375,
-0.037750244140625,
0.027313232421875,
0.0006403923034667969,
0.1021728515625,
0.0151824951171875,
-0.0316162109375,
0.007534027099609375,
-0.05224609375,
0.04937744140625,
-0.0240020751953125,
0.03466796875,
0.0063934326171875,
0.03558349609375,
0.01214599609375,
-0.08642578125,
-0.049713134765625,
0.00262451171875,
-0.01264190673828125,
0.0017461776733398438,
-0.017974853515625,
0.00820159912109375,
0.0282440185546875,
0.041229248046875,
-0.0132904052734375,
0.0030117034912109375,
-0.04815673828125,
-0.0362548828125,
0.033966064453125,
-0.00553131103515625,
0.00873565673828125,
-0.018157958984375,
-0.054046630859375,
-0.030364990234375,
-0.027313232421875,
0.020233154296875,
0.015411376953125,
0.01041412353515625,
-0.01049041748046875,
0.04193115234375,
0.0042266845703125,
0.04388427734375,
0.035675048828125,
-0.0023555755615234375,
0.0438232421875,
-0.021331787109375,
-0.0214691162109375,
-0.0010585784912109375,
0.06298828125,
0.024688720703125,
0.0182647705078125,
0.0005145072937011719,
-0.026123046875,
0.006565093994140625,
0.032867431640625,
-0.06689453125,
-0.0124664306640625,
-0.0143280029296875,
-0.0457763671875,
-0.032684326171875,
0.0125732421875,
-0.040802001953125,
-0.01251220703125,
-0.02557373046875,
0.0628662109375,
-0.01959228515625,
-0.0222930908203125,
-0.021484375,
0.0008115768432617188,
0.05145263671875,
0.017059326171875,
-0.06353759765625,
0.026702880859375,
0.025054931640625,
0.0701904296875,
-0.0009603500366210938,
-0.018280029296875,
-0.00417327880859375,
-0.01345062255859375,
-0.034454345703125,
0.0556640625,
-0.0181427001953125,
-0.0194549560546875,
0.00830078125,
0.034210205078125,
-0.00498199462890625,
-0.032623291015625,
0.03558349609375,
-0.0379638671875,
0.01354217529296875,
-0.0160369873046875,
-0.0235137939453125,
-0.0228729248046875,
0.014556884765625,
-0.05194091796875,
0.084228515625,
0.023651123046875,
-0.056182861328125,
0.036651611328125,
-0.044219970703125,
-0.006618499755859375,
0.00777435302734375,
-0.00608062744140625,
-0.04730224609375,
-0.0006809234619140625,
0.023040771484375,
0.041046142578125,
0.01110076904296875,
-0.00669097900390625,
-0.0225372314453125,
-0.034698486328125,
0.02398681640625,
-0.0172882080078125,
0.06109619140625,
0.0185394287109375,
-0.01531219482421875,
0.01287078857421875,
-0.049774169921875,
-0.00534820556640625,
0.0143280029296875,
-0.0201873779296875,
-0.00917816162109375,
-0.023101806640625,
0.0086822509765625,
0.025054931640625,
0.0340576171875,
-0.058837890625,
0.0179443359375,
-0.021087646484375,
0.04150390625,
0.06634521484375,
-0.00811004638671875,
0.0401611328125,
-0.013427734375,
0.0264739990234375,
0.0164947509765625,
0.04364013671875,
-0.03131103515625,
-0.039581298828125,
-0.06658935546875,
-0.0167236328125,
0.0302734375,
0.0251312255859375,
-0.054229736328125,
0.0396728515625,
-0.0252227783203125,
-0.0340576171875,
-0.04638671875,
0.00012153387069702148,
0.024658203125,
0.04150390625,
0.0277099609375,
-0.03765869140625,
-0.044036865234375,
-0.0723876953125,
0.0125579833984375,
0.0063934326171875,
0.0095977783203125,
0.01715087890625,
0.057830810546875,
-0.0267333984375,
0.07965087890625,
-0.0181884765625,
-0.01898193359375,
-0.00005906820297241211,
0.0018014907836914062,
0.03106689453125,
0.049102783203125,
0.05120849609375,
-0.0665283203125,
-0.0252838134765625,
-0.0007696151733398438,
-0.0634765625,
0.0180206298828125,
0.002483367919921875,
-0.0176849365234375,
0.00275421142578125,
0.01922607421875,
-0.05303955078125,
0.05963134765625,
0.0209197998046875,
-0.0194549560546875,
0.0234832763671875,
-0.0128173828125,
-0.00006157159805297852,
-0.08428955078125,
-0.00113677978515625,
0.0011615753173828125,
-0.034149169921875,
-0.04193115234375,
0.018463134765625,
0.01483917236328125,
-0.01399993896484375,
-0.0435791015625,
0.02813720703125,
-0.037353515625,
-0.028228759765625,
-0.0186767578125,
-0.02630615234375,
-0.0019063949584960938,
0.04547119140625,
0.0048370361328125,
0.03741455078125,
0.052825927734375,
-0.04180908203125,
0.048980712890625,
0.0279388427734375,
-0.0311126708984375,
0.03387451171875,
-0.050537109375,
0.0262603759765625,
-0.01045989990234375,
0.01477813720703125,
-0.059417724609375,
-0.0208892822265625,
0.020477294921875,
-0.032073974609375,
0.04296875,
-0.024932861328125,
-0.035003662109375,
-0.06298828125,
-0.019500732421875,
0.036651611328125,
0.050384521484375,
-0.0648193359375,
0.0526123046875,
0.0178375244140625,
0.0261383056640625,
-0.059906005859375,
-0.0677490234375,
-0.00653839111328125,
-0.0088653564453125,
-0.035247802734375,
0.04034423828125,
0.00594329833984375,
0.0212554931640625,
0.02667236328125,
0.0007138252258300781,
-0.016510009765625,
-0.019073486328125,
0.037017822265625,
0.0214996337890625,
-0.02166748046875,
0.0011510848999023438,
-0.0095977783203125,
-0.0134735107421875,
0.004428863525390625,
-0.03582763671875,
0.043212890625,
-0.034942626953125,
-0.0251312255859375,
-0.046478271484375,
0.0031490325927734375,
0.047149658203125,
-0.0189971923828125,
0.04644775390625,
0.058746337890625,
-0.05255126953125,
-0.0029926300048828125,
-0.027130126953125,
-0.0109100341796875,
-0.04144287109375,
0.0150299072265625,
-0.036376953125,
-0.04437255859375,
0.068115234375,
0.0033893585205078125,
-0.0186767578125,
0.045806884765625,
0.04193115234375,
-0.0177154541015625,
0.06097412109375,
0.05743408203125,
0.0012035369873046875,
0.05633544921875,
-0.0640869140625,
0.0136260986328125,
-0.061767578125,
-0.043212890625,
-0.00540924072265625,
-0.033447265625,
-0.035675048828125,
-0.0310821533203125,
0.0171661376953125,
0.0163116455078125,
-0.026214599609375,
0.047271728515625,
-0.053680419921875,
0.0305328369140625,
0.06365966796875,
0.044219970703125,
-0.0103759765625,
0.0035037994384765625,
-0.0223236083984375,
0.004589080810546875,
-0.042816162109375,
-0.0090484619140625,
0.076171875,
0.04046630859375,
0.06158447265625,
-0.01806640625,
0.050445556640625,
0.01403045654296875,
0.00920867919921875,
-0.0645751953125,
0.0396728515625,
-0.0026340484619140625,
-0.0509033203125,
-0.007114410400390625,
-0.00974273681640625,
-0.07159423828125,
0.0009813308715820312,
-0.0265960693359375,
-0.0516357421875,
0.05010986328125,
0.0201263427734375,
-0.0274505615234375,
0.033599853515625,
-0.04071044921875,
0.0687255859375,
-0.0223388671875,
-0.0218658447265625,
0.00516510009765625,
-0.04388427734375,
0.00836181640625,
-0.007320404052734375,
-0.01110076904296875,
0.021240234375,
0.028289794921875,
0.053985595703125,
-0.05316162109375,
0.07977294921875,
-0.0308837890625,
0.02276611328125,
0.042938232421875,
-0.01904296875,
0.0196990966796875,
-0.0185699462890625,
0.0278778076171875,
0.030120849609375,
-0.0002720355987548828,
-0.03326416015625,
-0.04046630859375,
0.033050537109375,
-0.07568359375,
-0.0307464599609375,
-0.0364990234375,
-0.0220947265625,
0.0199127197265625,
0.0298004150390625,
0.055633544921875,
0.04803466796875,
0.01446533203125,
0.0302886962890625,
0.04754638671875,
-0.020294189453125,
0.043212890625,
-0.004749298095703125,
-0.026824951171875,
-0.0187530517578125,
0.062225341796875,
0.0277862548828125,
0.005840301513671875,
0.0264434814453125,
0.019073486328125,
-0.03546142578125,
-0.0310821533203125,
-0.023284912109375,
0.00832366943359375,
-0.07080078125,
-0.032745361328125,
-0.032867431640625,
-0.05841064453125,
-0.03924560546875,
-0.01555633544921875,
-0.039337158203125,
-0.02288818359375,
-0.0355224609375,
-0.0234832763671875,
0.0292816162109375,
0.058197021484375,
-0.0222015380859375,
0.045867919921875,
-0.039031982421875,
0.01398468017578125,
0.052337646484375,
0.02783203125,
-0.0035114288330078125,
-0.05615234375,
-0.0285491943359375,
0.0016946792602539062,
-0.0183563232421875,
-0.05029296875,
0.03570556640625,
0.023651123046875,
0.06317138671875,
0.05548095703125,
-0.0167999267578125,
0.056793212890625,
-0.0264434814453125,
0.052337646484375,
0.0280914306640625,
-0.0640869140625,
0.0450439453125,
-0.01308441162109375,
0.00928497314453125,
0.01250457763671875,
0.033447265625,
-0.00992584228515625,
0.01039886474609375,
-0.046142578125,
-0.0472412109375,
0.046295166015625,
0.006374359130859375,
0.0261383056640625,
0.013519287109375,
0.043182373046875,
-0.0006437301635742188,
0.004852294921875,
-0.077392578125,
-0.0120849609375,
-0.06781005859375,
-0.01377105712890625,
0.01526641845703125,
-0.019256591796875,
-0.00203704833984375,
-0.046295166015625,
0.0178375244140625,
-0.0120697021484375,
0.0670166015625,
0.0194244384765625,
-0.01971435546875,
-0.005481719970703125,
-0.02545166015625,
0.0162506103515625,
0.03436279296875,
-0.0295562744140625,
0.01103973388671875,
0.004302978515625,
-0.043487548828125,
-0.004924774169921875,
0.001041412353515625,
-0.011810302734375,
-0.00945281982421875,
0.034576416015625,
0.077392578125,
0.0136566162109375,
0.0009832382202148438,
0.0660400390625,
0.01145172119140625,
-0.023162841796875,
-0.041748046875,
0.004985809326171875,
-0.0189361572265625,
0.03363037109375,
0.03558349609375,
0.0275726318359375,
0.00421142578125,
-0.04449462890625,
0.0244293212890625,
0.01605224609375,
-0.0426025390625,
-0.044830322265625,
0.058319091796875,
-0.005619049072265625,
-0.01016998291015625,
0.05181884765625,
-0.00894927978515625,
-0.056610107421875,
0.05535888671875,
0.048126220703125,
0.059234619140625,
-0.023712158203125,
0.01265716552734375,
0.035186767578125,
0.0216522216796875,
-0.0018711090087890625,
0.017669677734375,
-0.01172637939453125,
-0.0845947265625,
-0.0279693603515625,
-0.05133056640625,
-0.0094451904296875,
0.01155853271484375,
-0.0582275390625,
0.0204315185546875,
-0.046783447265625,
-0.028594970703125,
0.0142974853515625,
-0.01316070556640625,
-0.08416748046875,
0.02630615234375,
0.037109375,
0.058837890625,
-0.0614013671875,
0.0762939453125,
0.053131103515625,
-0.044830322265625,
-0.056793212890625,
-0.03472900390625,
-0.017486572265625,
-0.07708740234375,
0.06378173828125,
0.0264739990234375,
0.0016183853149414062,
0.0029754638671875,
-0.06915283203125,
-0.0748291015625,
0.08660888671875,
0.0247039794921875,
-0.021942138671875,
-0.0085296630859375,
0.0033969879150390625,
0.039398193359375,
-0.0404052734375,
0.0260009765625,
-0.0009489059448242188,
0.014678955078125,
0.0252227783203125,
-0.057891845703125,
-0.00594329833984375,
-0.031951904296875,
0.0181121826171875,
-0.007472991943359375,
-0.0567626953125,
0.08172607421875,
-0.01451873779296875,
-0.013641357421875,
0.0021820068359375,
0.05108642578125,
-0.0142974853515625,
0.006191253662109375,
0.048675537109375,
0.05218505859375,
0.0416259765625,
-0.0196685791015625,
0.07525634765625,
-0.01151275634765625,
0.0479736328125,
0.048187255859375,
0.0102691650390625,
0.05072021484375,
0.0203704833984375,
-0.01473236083984375,
0.04937744140625,
0.06939697265625,
-0.039031982421875,
0.061126708984375,
-0.0009717941284179688,
0.006103515625,
-0.01580810546875,
0.00286865234375,
-0.0288543701171875,
0.04779052734375,
0.035186767578125,
-0.048736572265625,
0.0031948089599609375,
0.0260467529296875,
-0.0241546630859375,
-0.020660400390625,
-0.04534912109375,
0.03656005859375,
0.006534576416015625,
-0.021087646484375,
0.044647216796875,
-0.0227203369140625,
0.043975830078125,
-0.0276947021484375,
-0.00579833984375,
-0.01480865478515625,
0.02374267578125,
-0.0265960693359375,
-0.06378173828125,
0.012176513671875,
-0.0162200927734375,
-0.00943756103515625,
-0.0157623291015625,
0.0687255859375,
-0.0120697021484375,
-0.045806884765625,
0.027130126953125,
0.005828857421875,
0.0206451416015625,
0.00659942626953125,
-0.053314208984375,
-0.0178070068359375,
-0.0164337158203125,
-0.028564453125,
0.01172637939453125,
0.0265655517578125,
-0.006805419921875,
0.04949951171875,
0.049591064453125,
-0.00774383544921875,
0.04132080078125,
-0.0013093948364257812,
0.08697509765625,
-0.0491943359375,
-0.040252685546875,
-0.04132080078125,
0.039398193359375,
-0.0167388916015625,
-0.0215301513671875,
0.03985595703125,
0.033599853515625,
0.078369140625,
-0.0172271728515625,
0.036834716796875,
-0.006999969482421875,
0.0173797607421875,
-0.0276336669921875,
0.04559326171875,
-0.03289794921875,
-0.0164642333984375,
-0.0103607177734375,
-0.078857421875,
-0.016021728515625,
0.07086181640625,
-0.0027637481689453125,
0.01070404052734375,
0.036285400390625,
0.052001953125,
-0.0221099853515625,
-0.0178070068359375,
0.02728271484375,
0.0304718017578125,
0.00125885009765625,
0.0257110595703125,
0.06927490234375,
-0.04644775390625,
0.03485107421875,
-0.0404052734375,
-0.01497650146484375,
-0.01207733154296875,
-0.050628662109375,
-0.08074951171875,
-0.04986572265625,
-0.0225067138671875,
-0.032501220703125,
-0.006855010986328125,
0.05352783203125,
0.08489990234375,
-0.06866455078125,
0.00786590576171875,
0.00009942054748535156,
-0.01203155517578125,
-0.0166015625,
-0.014495849609375,
0.03533935546875,
0.0038738250732421875,
-0.05419921875,
0.004146575927734375,
0.00273895263671875,
0.019195556640625,
-0.031463623046875,
0.0015697479248046875,
-0.001468658447265625,
-0.01373291015625,
0.043060302734375,
0.025543212890625,
-0.04852294921875,
-0.045867919921875,
-0.00037479400634765625,
-0.002178192138671875,
0.026947021484375,
0.0382080078125,
-0.07342529296875,
0.042938232421875,
0.03179931640625,
0.03631591796875,
0.07720947265625,
-0.0038604736328125,
0.02099609375,
-0.05902099609375,
0.0202789306640625,
0.004535675048828125,
0.051422119140625,
0.0226287841796875,
-0.035308837890625,
0.03338623046875,
0.035736083984375,
-0.035614013671875,
-0.055877685546875,
0.0080108642578125,
-0.09576416015625,
-0.00555419921875,
0.06036376953125,
-0.031768798828125,
-0.041412353515625,
0.012359619140625,
-0.004749298095703125,
0.038299560546875,
-0.0097808837890625,
0.040008544921875,
0.0264739990234375,
-0.0006551742553710938,
-0.0455322265625,
-0.020904541015625,
0.01837158203125,
-0.01934814453125,
-0.034942626953125,
-0.049530029296875,
-0.00113677978515625,
0.0233306884765625,
0.04248046875,
0.026824951171875,
-0.0284881591796875,
0.007114410400390625,
0.01873779296875,
0.0286865234375,
-0.01300811767578125,
-0.02716064453125,
-0.0244293212890625,
0.0035152435302734375,
-0.0224609375,
-0.054168701171875
]
] |
nlp-waseda/roberta-base-japanese | 2022-10-21T14:46:36.000Z | [
"transformers",
"pytorch",
"roberta",
"fill-mask",
"ja",
"dataset:wikipedia",
"dataset:cc100",
"license:cc-by-sa-4.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | fill-mask | nlp-waseda | null | null | nlp-waseda/roberta-base-japanese | 22 | 7,318 | transformers | 2022-03-02T23:29:05 | ---
language: ja
license: cc-by-sa-4.0
datasets:
- wikipedia
- cc100
mask_token: "[MASK]"
widget:
- text: "早稲田 大学 で 自然 言語 処理 を [MASK] する 。"
---
# nlp-waseda/roberta-base-japanese
## Model description
This is a Japanese RoBERTa base model pretrained on Japanese Wikipedia and the Japanese portion of CC-100.
## How to use
You can use this model for masked language modeling as follows:
```python
from transformers import AutoTokenizer, AutoModelForMaskedLM
tokenizer = AutoTokenizer.from_pretrained("nlp-waseda/roberta-base-japanese")
model = AutoModelForMaskedLM.from_pretrained("nlp-waseda/roberta-base-japanese")
sentence = '早稲田 大学 で 自然 言語 処理 を [MASK] する 。' # input should be segmented into words by Juman++ in advance
encoding = tokenizer(sentence, return_tensors='pt')
...
```
You can fine-tune this model on downstream tasks.
## Tokenization
The input text should be segmented into words by [Juman++](https://github.com/ku-nlp/jumanpp) in advance. Juman++ 2.0.0-rc3 was used for pretraining. Each word is tokenized into tokens by [sentencepiece](https://github.com/google/sentencepiece).
`BertJapaneseTokenizer` now supports automatic `JumanppTokenizer` and `SentencepieceTokenizer`. You can use [this model](https://huggingface.co/nlp-waseda/roberta-base-japanese-with-auto-jumanpp) without any data preprocessing.
## Vocabulary
The vocabulary consists of 32000 tokens including words ([JumanDIC](https://github.com/ku-nlp/JumanDIC)) and subwords induced by the unigram language model of [sentencepiece](https://github.com/google/sentencepiece).
## Training procedure
This model was trained on Japanese Wikipedia (as of 20210920) and the Japanese portion of CC-100. It took a week using eight NVIDIA A100 GPUs.
The following hyperparameters were used during pretraining:
- learning_rate: 1e-4
- per_device_train_batch_size: 256
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 2
- total_train_batch_size: 4096
- max_seq_length: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 700000
- warmup_steps: 10000
- mixed_precision_training: Native AMP
## Performance on JGLUE
See the [Baseline Scores](https://github.com/yahoojapan/JGLUE#baseline-scores) of JGLUE.
| 2,270 | [
[
-0.03326416015625,
-0.06793212890625,
0.01526641845703125,
0.0161895751953125,
-0.03759765625,
0.001415252685546875,
-0.033233642578125,
-0.022216796875,
0.033660888671875,
0.047210693359375,
-0.0604248046875,
-0.03924560546875,
-0.049713134765625,
0.004009246826171875,
-0.01166534423828125,
0.0863037109375,
0.004337310791015625,
0.021728515625,
0.0132598876953125,
0.0017185211181640625,
-0.039337158203125,
-0.039276123046875,
-0.0562744140625,
-0.03851318359375,
0.01053619384765625,
0.0007815361022949219,
0.041351318359375,
0.03753662109375,
0.0190277099609375,
0.025787353515625,
-0.0131683349609375,
0.0076446533203125,
-0.009063720703125,
-0.005840301513671875,
-0.0009713172912597656,
-0.04052734375,
-0.039581298828125,
0.0018873214721679688,
0.041015625,
0.044281005859375,
-0.00704193115234375,
0.0098876953125,
0.004337310791015625,
0.0362548828125,
-0.046661376953125,
0.020263671875,
-0.045166015625,
0.0228424072265625,
-0.037078857421875,
-0.002262115478515625,
-0.0269012451171875,
0.01073455810546875,
0.006072998046875,
-0.0693359375,
0.0107421875,
0.0048065185546875,
0.0946044921875,
0.021270751953125,
-0.0011148452758789062,
-0.02410888671875,
-0.0328369140625,
0.060394287109375,
-0.061981201171875,
0.0121917724609375,
0.046905517578125,
0.0111083984375,
-0.00872802734375,
-0.055694580078125,
-0.045379638671875,
-0.016143798828125,
-0.00965118408203125,
0.029052734375,
-0.006984710693359375,
0.0005316734313964844,
0.04974365234375,
0.0307769775390625,
-0.061126708984375,
0.00208282470703125,
-0.0236053466796875,
-0.01690673828125,
0.038787841796875,
0.0033168792724609375,
0.03045654296875,
-0.039947509765625,
-0.0250396728515625,
-0.0181121826171875,
-0.03497314453125,
-0.0048370361328125,
0.035369873046875,
0.0228729248046875,
-0.02679443359375,
0.037994384765625,
-0.0249176025390625,
0.044281005859375,
0.0031032562255859375,
-0.017791748046875,
0.0311737060546875,
-0.01497650146484375,
-0.0207672119140625,
-0.010040283203125,
0.08123779296875,
-0.0020923614501953125,
0.0304107666015625,
-0.004505157470703125,
-0.004375457763671875,
0.0032520294189453125,
0.021728515625,
-0.05810546875,
-0.026458740234375,
0.0026092529296875,
-0.019866943359375,
-0.0133514404296875,
-0.0041656494140625,
-0.047393798828125,
0.010772705078125,
-0.022308349609375,
0.054229736328125,
-0.041595458984375,
-0.0095062255859375,
0.0250091552734375,
-0.00994110107421875,
0.00836181640625,
0.005863189697265625,
-0.07427978515625,
0.034576416015625,
0.033355712890625,
0.0595703125,
-0.01282501220703125,
-0.0226287841796875,
-0.016082763671875,
-0.006595611572265625,
-0.01438140869140625,
0.02978515625,
-0.00582122802734375,
-0.030303955078125,
-0.01107025146484375,
0.018463134765625,
-0.0207672119140625,
-0.0226898193359375,
0.04119873046875,
-0.0302734375,
0.06573486328125,
0.005558013916015625,
-0.053253173828125,
-0.01800537109375,
0.0235443115234375,
-0.040313720703125,
0.0797119140625,
0.01605224609375,
-0.059417724609375,
0.0265045166015625,
-0.058837890625,
-0.0225067138671875,
0.03631591796875,
-0.00576019287109375,
-0.0411376953125,
-0.0019359588623046875,
0.033905029296875,
0.0254669189453125,
-0.0024738311767578125,
0.032745361328125,
-0.005062103271484375,
-0.035003662109375,
0.0126495361328125,
-0.034912109375,
0.09637451171875,
0.00533294677734375,
-0.044189453125,
0.0008673667907714844,
-0.06427001953125,
0.0089874267578125,
0.0103912353515625,
-0.01934814453125,
-0.0253143310546875,
-0.027679443359375,
0.0197906494140625,
0.00801849365234375,
0.0169219970703125,
-0.046661376953125,
0.0115814208984375,
-0.051177978515625,
0.02899169921875,
0.035858154296875,
-0.010223388671875,
0.020294189453125,
-0.0030040740966796875,
0.0562744140625,
-0.00461578369140625,
0.0218048095703125,
-0.01416015625,
-0.042327880859375,
-0.08416748046875,
-0.023956298828125,
0.043609619140625,
0.045867919921875,
-0.06756591796875,
0.06732177734375,
-0.01461029052734375,
-0.032501220703125,
-0.045440673828125,
-0.00862884521484375,
0.052459716796875,
0.035064697265625,
0.037384033203125,
-0.051422119140625,
-0.04888916015625,
-0.05865478515625,
0.0023059844970703125,
0.0022945404052734375,
-0.004421234130859375,
0.01129150390625,
0.053192138671875,
-0.0172882080078125,
0.045867919921875,
-0.042724609375,
-0.0301055908203125,
-0.0169830322265625,
0.0190582275390625,
0.029327392578125,
0.05377197265625,
0.0271759033203125,
-0.03619384765625,
-0.032501220703125,
-0.019989013671875,
-0.04962158203125,
0.00589752197265625,
0.00201416015625,
-0.0203399658203125,
0.0282135009765625,
0.04071044921875,
-0.05084228515625,
0.03289794921875,
0.049957275390625,
-0.01904296875,
0.04266357421875,
-0.01617431640625,
-0.00909423828125,
-0.1097412109375,
0.02777099609375,
-0.01262664794921875,
-0.02276611328125,
-0.038970947265625,
0.0273590087890625,
-0.0038852691650390625,
-0.0247344970703125,
-0.03546142578125,
0.043701171875,
-0.030364990234375,
-0.002330780029296875,
-0.034759521484375,
-0.00638580322265625,
0.0016260147094726562,
0.06524658203125,
0.028350830078125,
0.061279296875,
0.044677734375,
-0.035491943359375,
0.00949859619140625,
0.0308837890625,
-0.04388427734375,
0.01503753662109375,
-0.06805419921875,
0.00594329833984375,
0.01386260986328125,
0.01229095458984375,
-0.049285888671875,
-0.0241546630859375,
0.043060302734375,
-0.037628173828125,
0.0304107666015625,
-0.025238037109375,
-0.043487548828125,
-0.028656005859375,
0.006534576416015625,
0.024169921875,
0.046356201171875,
-0.0233306884765625,
0.044036865234375,
0.0245208740234375,
-0.009063720703125,
-0.054718017578125,
-0.058563232421875,
0.01605224609375,
-0.0153350830078125,
-0.0283355712890625,
0.049468994140625,
-0.01410675048828125,
0.0166473388671875,
-0.0026092529296875,
0.01326751708984375,
-0.0153045654296875,
0.0191497802734375,
0.0213470458984375,
0.034393310546875,
-0.006580352783203125,
-0.0048675537109375,
-0.00028705596923828125,
-0.0291900634765625,
0.006809234619140625,
-0.00815582275390625,
0.08197021484375,
-0.007633209228515625,
-0.00954437255859375,
-0.037078857421875,
0.003528594970703125,
0.015228271484375,
-0.0129547119140625,
0.0609130859375,
0.0693359375,
-0.0262298583984375,
-0.0090484619140625,
-0.027099609375,
-0.005565643310546875,
-0.03485107421875,
0.033172607421875,
-0.0347900390625,
-0.057830810546875,
0.033477783203125,
0.015838623046875,
-0.00879669189453125,
0.046722412109375,
0.0465087890625,
0.006252288818359375,
0.07806396484375,
0.04180908203125,
-0.0133514404296875,
0.036407470703125,
-0.035064697265625,
-0.0037136077880859375,
-0.07080078125,
-0.01328277587890625,
-0.0237579345703125,
0.0053558349609375,
-0.058349609375,
-0.02020263671875,
0.027008056640625,
-0.0034084320068359375,
-0.0122833251953125,
0.04034423828125,
-0.0374755859375,
0.04345703125,
0.056060791015625,
0.01393890380859375,
0.0113372802734375,
0.0002961158752441406,
-0.0120849609375,
-0.00572967529296875,
-0.05999755859375,
-0.0261383056640625,
0.08319091796875,
0.0292205810546875,
0.040496826171875,
-0.0168914794921875,
0.04827880859375,
-0.00412750244140625,
0.00008189678192138672,
-0.051422119140625,
0.0362548828125,
-0.03228759765625,
-0.06976318359375,
-0.015228271484375,
-0.01763916015625,
-0.06414794921875,
0.0267486572265625,
-0.032562255859375,
-0.037078857421875,
-0.01525115966796875,
-0.0022296905517578125,
-0.0158843994140625,
0.0040130615234375,
-0.0273284912109375,
0.07177734375,
-0.01442718505859375,
0.00024771690368652344,
-0.007686614990234375,
-0.057647705078125,
0.0408935546875,
-0.003971099853515625,
0.0020542144775390625,
-0.006160736083984375,
0.0193939208984375,
0.06744384765625,
-0.034332275390625,
0.052215576171875,
-0.01267242431640625,
0.028472900390625,
-0.004856109619140625,
-0.01617431640625,
0.008392333984375,
0.0024394989013671875,
0.0194244384765625,
0.0321044921875,
-0.00255584716796875,
-0.031890869140625,
-0.029296875,
0.037506103515625,
-0.07989501953125,
-0.0298309326171875,
-0.049896240234375,
-0.038848876953125,
-0.0032825469970703125,
0.0236663818359375,
0.053619384765625,
0.033660888671875,
-0.005084991455078125,
0.0186004638671875,
0.051422119140625,
-0.0191802978515625,
0.03863525390625,
0.03802490234375,
-0.025604248046875,
-0.043426513671875,
0.049285888671875,
0.01076507568359375,
0.0082244873046875,
0.0208740234375,
0.0179290771484375,
-0.01538848876953125,
-0.0438232421875,
-0.054229736328125,
0.027984619140625,
-0.0292816162109375,
-0.0002770423889160156,
-0.055145263671875,
-0.040679931640625,
-0.019256591796875,
0.009765625,
-0.033721923828125,
-0.035400390625,
-0.022491455078125,
0.00852203369140625,
0.00710296630859375,
0.0197906494140625,
0.01397705078125,
0.043487548828125,
-0.037017822265625,
0.0157470703125,
-0.0010137557983398438,
0.011566162109375,
-0.01458740234375,
-0.06732177734375,
-0.0343017578125,
0.0047149658203125,
-0.01270294189453125,
-0.038421630859375,
0.05615234375,
0.003467559814453125,
0.0308380126953125,
0.012725830078125,
-0.0162811279296875,
0.072265625,
-0.0404052734375,
0.064208984375,
0.01410675048828125,
-0.08184814453125,
0.04248046875,
-0.022552490234375,
0.04296875,
0.053375244140625,
0.04156494140625,
-0.03497314453125,
-0.035858154296875,
-0.06683349609375,
-0.0692138671875,
0.0574951171875,
0.0269622802734375,
0.0192108154296875,
0.0023670196533203125,
0.023590087890625,
0.003704071044921875,
0.019378662109375,
-0.09271240234375,
-0.0247344970703125,
-0.0313720703125,
-0.01207733154296875,
-0.023681640625,
-0.00591278076171875,
-0.00614166259765625,
-0.0257568359375,
0.08001708984375,
-0.0018854141235351562,
0.01800537109375,
0.004207611083984375,
-0.03192138671875,
0.0086669921875,
0.00145721435546875,
0.0506591796875,
0.032745361328125,
-0.0251922607421875,
-0.0168304443359375,
0.01056671142578125,
-0.046630859375,
-0.00881195068359375,
0.0171661376953125,
-0.0380859375,
0.0265350341796875,
0.0361328125,
0.0830078125,
0.01195526123046875,
-0.039337158203125,
0.040191650390625,
-0.010345458984375,
-0.01910400390625,
-0.061767578125,
0.015228271484375,
-0.00685882568359375,
-0.00827789306640625,
0.01474761962890625,
-0.0086517333984375,
0.0022716522216796875,
-0.038421630859375,
0.00689697265625,
0.0308380126953125,
-0.0256805419921875,
-0.0191802978515625,
0.055328369140625,
0.006519317626953125,
-0.0239105224609375,
0.06353759765625,
-0.0229339599609375,
-0.049468994140625,
0.0374755859375,
0.045684814453125,
0.06866455078125,
-0.0036792755126953125,
0.01116180419921875,
0.050506591796875,
0.0111083984375,
0.0003185272216796875,
0.00684356689453125,
-0.00104522705078125,
-0.062225341796875,
-0.043914794921875,
-0.065185546875,
-0.0240325927734375,
0.035888671875,
-0.061004638671875,
0.0156402587890625,
-0.04486083984375,
-0.0213775634765625,
0.00644683837890625,
0.030670166015625,
-0.04730224609375,
0.031341552734375,
0.00913238525390625,
0.062103271484375,
-0.07623291015625,
0.07373046875,
0.04119873046875,
-0.047393798828125,
-0.0662841796875,
-0.01397705078125,
-0.01309967041015625,
-0.08465576171875,
0.05682373046875,
0.015838623046875,
0.020538330078125,
0.00557708740234375,
-0.031951904296875,
-0.068359375,
0.0753173828125,
-0.003589630126953125,
-0.044647216796875,
-0.0056915283203125,
0.00615692138671875,
0.05108642578125,
-0.0304107666015625,
0.02740478515625,
0.02252197265625,
0.014129638671875,
-0.0021076202392578125,
-0.06494140625,
-0.0012254714965820312,
-0.0298309326171875,
0.01319122314453125,
0.0118865966796875,
-0.035888671875,
0.071044921875,
0.027435302734375,
-0.005222320556640625,
0.0269012451171875,
0.0386962890625,
0.024688720703125,
0.0010080337524414062,
0.0390625,
0.07177734375,
0.038665771484375,
-0.01352691650390625,
0.0718994140625,
-0.035400390625,
0.0273284912109375,
0.076904296875,
0.01080322265625,
0.0576171875,
0.019775390625,
-0.0163421630859375,
0.060211181640625,
0.03887939453125,
-0.019805908203125,
0.040985107421875,
0.0118865966796875,
-0.01513671875,
-0.01056671142578125,
-0.0026798248291015625,
-0.024444580078125,
0.053131103515625,
0.0212860107421875,
-0.0279541015625,
0.004425048828125,
0.0172271728515625,
0.0306396484375,
-0.01000213623046875,
-0.0140228271484375,
0.058685302734375,
-0.0002117156982421875,
-0.0638427734375,
0.036651611328125,
0.020751953125,
0.0677490234375,
-0.06610107421875,
0.018768310546875,
-0.01430511474609375,
0.005886077880859375,
0.0035381317138671875,
-0.036956787109375,
0.022674560546875,
0.01068115234375,
-0.0095977783203125,
-0.0146942138671875,
0.0384521484375,
-0.047607421875,
-0.04974365234375,
0.005619049072265625,
0.0275421142578125,
0.024261474609375,
-0.0014896392822265625,
-0.07659912109375,
0.006633758544921875,
0.002429962158203125,
-0.022979736328125,
0.027587890625,
0.0007772445678710938,
0.0015201568603515625,
0.04058837890625,
0.0565185546875,
0.01389312744140625,
0.002750396728515625,
0.02447509765625,
0.05059814453125,
-0.040924072265625,
-0.039794921875,
-0.06134033203125,
0.03857421875,
-0.0123443603515625,
-0.050018310546875,
0.06195068359375,
0.043487548828125,
0.07281494140625,
-0.0168914794921875,
0.039520263671875,
-0.0074462890625,
0.035858154296875,
-0.06817626953125,
0.053253173828125,
-0.035858154296875,
-0.0015115737915039062,
-0.02215576171875,
-0.07171630859375,
-0.018310546875,
0.064208984375,
-0.002834320068359375,
0.004299163818359375,
0.04962158203125,
0.058868408203125,
-0.0009136199951171875,
-0.016815185546875,
0.0200958251953125,
0.01119232177734375,
0.01221466064453125,
0.0413818359375,
0.042877197265625,
-0.05413818359375,
0.0171661376953125,
-0.0421142578125,
-0.0099334716796875,
-0.0040435791015625,
-0.062103271484375,
-0.076171875,
-0.04290771484375,
-0.0382080078125,
-0.023193359375,
0.01445770263671875,
0.0711669921875,
0.068603515625,
-0.056365966796875,
-0.0278778076171875,
-0.0258941650390625,
-0.0259246826171875,
0.0033740997314453125,
-0.0188446044921875,
0.033660888671875,
-0.034515380859375,
-0.07427978515625,
0.0224151611328125,
-0.0012845993041992188,
0.00815582275390625,
-0.02850341796875,
0.0020694732666015625,
-0.0160675048828125,
-0.0167083740234375,
0.033416748046875,
0.00952911376953125,
-0.06060791015625,
-0.00740814208984375,
-0.0028972625732421875,
-0.00675201416015625,
-0.0072479248046875,
0.030975341796875,
-0.053985595703125,
0.0411376953125,
0.0155181884765625,
0.032562255859375,
0.07952880859375,
-0.0019273757934570312,
0.040313720703125,
-0.06781005859375,
0.0309906005859375,
0.01393890380859375,
0.036376953125,
0.026092529296875,
-0.00537109375,
0.04534912109375,
0.0361328125,
-0.04022216796875,
-0.048126220703125,
0.011383056640625,
-0.08306884765625,
-0.034088134765625,
0.06915283203125,
-0.02008056640625,
-0.024261474609375,
0.00621795654296875,
-0.01024627685546875,
0.0277862548828125,
0.0028018951416015625,
0.056060791015625,
0.0660400390625,
0.01285552978515625,
-0.01183319091796875,
-0.03546142578125,
0.038970947265625,
0.041595458984375,
-0.0562744140625,
-0.0308380126953125,
0.0220489501953125,
0.03363037109375,
0.0229949951171875,
0.04913330078125,
-0.01146697998046875,
0.0137939453125,
0.002872467041015625,
0.0243377685546875,
-0.017303466796875,
-0.02001953125,
-0.0296630859375,
-0.005512237548828125,
-0.01554107666015625,
-0.025054931640625
]
] |
Narsil/deberta-large-mnli-zero-cls | 2021-08-23T13:27:24.000Z | [
"transformers",
"pytorch",
"deberta",
"text-classification",
"deberta-v1",
"deberta-mnli",
"zero-shot-classification",
"en",
"arxiv:2006.03654",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | zero-shot-classification | Narsil | null | null | Narsil/deberta-large-mnli-zero-cls | 11 | 7,312 | transformers | 2022-03-02T23:29:04 | ---
language: en
tags:
- deberta-v1
- deberta-mnli
tasks: mnli
thumbnail: https://huggingface.co/front/thumbnails/microsoft.png
license: mit
pipeline_tag: zero-shot-classification
---
## DeBERTa: Decoding-enhanced BERT with Disentangled Attention
[DeBERTa](https://arxiv.org/abs/2006.03654) improves the BERT and RoBERTa models using disentangled attention and enhanced mask decoder. It outperforms BERT and RoBERTa on majority of NLU tasks with 80GB training data.
Please check the [official repository](https://github.com/microsoft/DeBERTa) for more details and updates.
This is the DeBERTa large model fine-tuned with MNLI task.
#### Fine-tuning on NLU tasks
We present the dev results on SQuAD 1.1/2.0 and several GLUE benchmark tasks.
| Model | SQuAD 1.1 | SQuAD 2.0 | MNLI-m/mm | SST-2 | QNLI | CoLA | RTE | MRPC | QQP |STS-B |
|---------------------------|-----------|-----------|-------------|-------|------|------|--------|-------|-------|------|
| | F1/EM | F1/EM | Acc | Acc | Acc | MCC | Acc |Acc/F1 |Acc/F1 |P/S |
| BERT-Large | 90.9/84.1 | 81.8/79.0 | 86.6/- | 93.2 | 92.3 | 60.6 | 70.4 | 88.0/- | 91.3/- |90.0/- |
| RoBERTa-Large | 94.6/88.9 | 89.4/86.5 | 90.2/- | 96.4 | 93.9 | 68.0 | 86.6 | 90.9/- | 92.2/- |92.4/- |
| XLNet-Large | 95.1/89.7 | 90.6/87.9 | 90.8/- | 97.0 | 94.9 | 69.0 | 85.9 | 90.8/- | 92.3/- |92.5/- |
| [DeBERTa-Large](https://huggingface.co/microsoft/deberta-large)<sup>1</sup> | 95.5/90.1 | 90.7/88.0 | 91.3/91.1| 96.5|95.3| 69.5| 91.0| 92.6/94.6| 92.3/- |92.8/92.5 |
| [DeBERTa-XLarge](https://huggingface.co/microsoft/deberta-xlarge)<sup>1</sup> | -/- | -/- | 91.5/91.2| 97.0 | - | - | 93.1 | 92.1/94.3 | - |92.9/92.7|
| [DeBERTa-V2-XLarge](https://huggingface.co/microsoft/deberta-v2-xlarge)<sup>1</sup>|95.8/90.8| 91.4/88.9|91.7/91.6| **97.5**| 95.8|71.1|**93.9**|92.0/94.2|92.3/89.8|92.9/92.9|
|**[DeBERTa-V2-XXLarge](https://huggingface.co/microsoft/deberta-v2-xxlarge)<sup>1,2</sup>**|**96.1/91.4**|**92.2/89.7**|**91.7/91.9**|97.2|**96.0**|**72.0**| 93.5| **93.1/94.9**|**92.7/90.3** |**93.2/93.1** |
--------
#### Notes.
- <sup>1</sup> Following RoBERTa, for RTE, MRPC, STS-B, we fine-tune the tasks based on [DeBERTa-Large-MNLI](https://huggingface.co/microsoft/deberta-large-mnli), [DeBERTa-XLarge-MNLI](https://huggingface.co/microsoft/deberta-xlarge-mnli), [DeBERTa-V2-XLarge-MNLI](https://huggingface.co/microsoft/deberta-v2-xlarge-mnli), [DeBERTa-V2-XXLarge-MNLI](https://huggingface.co/microsoft/deberta-v2-xxlarge-mnli). The results of SST-2/QQP/QNLI/SQuADv2 will also be slightly improved when start from MNLI fine-tuned models, however, we only report the numbers fine-tuned from pretrained base models for those 4 tasks.
- <sup>2</sup> To try the **XXLarge** model with **[HF transformers](https://huggingface.co/transformers/main_classes/trainer.html)**, you need to specify **--sharded_ddp**
```bash
cd transformers/examples/text-classification/
export TASK_NAME=mrpc
python -m torch.distributed.launch --nproc_per_node=8 run_glue.py --model_name_or_path microsoft/deberta-v2-xxlarge \\\n--task_name $TASK_NAME --do_train --do_eval --max_seq_length 128 --per_device_train_batch_size 4 \\\n--learning_rate 3e-6 --num_train_epochs 3 --output_dir /tmp/$TASK_NAME/ --overwrite_output_dir --sharded_ddp --fp16
```
### Citation
If you find DeBERTa useful for your work, please cite the following paper:
``` latex
@inproceedings{
he2021deberta,
title={DEBERTA: DECODING-ENHANCED BERT WITH DISENTANGLED ATTENTION},
author={Pengcheng He and Xiaodong Liu and Jianfeng Gao and Weizhu Chen},
booktitle={International Conference on Learning Representations},
year={2021},
url={https://openreview.net/forum?id=XPZIaotutsD}
}
```
| 3,888 | [
[
-0.034149169921875,
-0.0467529296875,
0.020843505859375,
0.034698486328125,
-0.01325225830078125,
0.01313018798828125,
-0.0005970001220703125,
-0.048095703125,
0.02191162109375,
0.0136260986328125,
-0.061798095703125,
-0.0255584716796875,
-0.06878662109375,
-0.005275726318359375,
-0.0007295608520507812,
0.06475830078125,
-0.004871368408203125,
-0.01438140869140625,
-0.00991058349609375,
-0.01371002197265625,
-0.043304443359375,
-0.034454345703125,
-0.0394287109375,
-0.032806396484375,
0.0231475830078125,
0.0245208740234375,
0.05047607421875,
0.00991058349609375,
0.03692626953125,
0.0255584716796875,
-0.0316162109375,
0.0274658203125,
-0.038238525390625,
-0.004917144775390625,
0.01033782958984375,
-0.025299072265625,
-0.0693359375,
0.00693511962890625,
0.026458740234375,
0.02752685546875,
0.0171051025390625,
0.025543212890625,
0.029632568359375,
0.07879638671875,
-0.036865234375,
0.009552001953125,
-0.035797119140625,
0.00550079345703125,
0.00963592529296875,
-0.0031528472900390625,
-0.01503753662109375,
-0.00293731689453125,
0.005084991455078125,
-0.029541015625,
0.0016660690307617188,
-0.01329803466796875,
0.09356689453125,
0.04010009765625,
-0.0101470947265625,
-0.00791168212890625,
-0.0321044921875,
0.0858154296875,
-0.055389404296875,
0.0261688232421875,
0.02508544921875,
0.00304412841796875,
-0.01543426513671875,
-0.0308990478515625,
-0.0287017822265625,
-0.0130157470703125,
-0.01500701904296875,
0.022430419921875,
-0.05596923828125,
-0.01031494140625,
0.0313720703125,
0.01139068603515625,
-0.051513671875,
0.0140533447265625,
-0.02593994140625,
0.0009794235229492188,
0.054962158203125,
0.006427764892578125,
0.013946533203125,
0.00748443603515625,
-0.0372314453125,
-0.00989532470703125,
-0.042816162109375,
0.016876220703125,
0.0098724365234375,
0.0016031265258789062,
-0.0236663818359375,
0.0196075439453125,
-0.01910400390625,
0.06817626953125,
0.0299835205078125,
-0.0014390945434570312,
0.05426025390625,
-0.014068603515625,
-0.0345458984375,
0.0011129379272460938,
0.049346923828125,
0.0248260498046875,
-0.0018320083618164062,
-0.00681304931640625,
-0.0162353515625,
0.003650665283203125,
0.006153106689453125,
-0.072509765625,
-0.032012939453125,
0.0411376953125,
-0.045135498046875,
-0.0188446044921875,
-0.004039764404296875,
-0.040313720703125,
-0.0003032684326171875,
-0.046142578125,
0.0288543701171875,
-0.04217529296875,
-0.02435302734375,
0.003482818603515625,
-0.0035991668701171875,
0.004711151123046875,
0.037353515625,
-0.0643310546875,
0.0003666877746582031,
0.0355224609375,
0.055419921875,
-0.00714874267578125,
-0.0136260986328125,
-0.045806884765625,
-0.014862060546875,
-0.004474639892578125,
0.0204315185546875,
-0.0134735107421875,
0.0082244873046875,
-0.01336669921875,
0.01308441162109375,
-0.0247802734375,
-0.026885986328125,
0.01432037353515625,
-0.039703369140625,
-0.002918243408203125,
-0.0290679931640625,
-0.0284576416015625,
-0.0201263427734375,
0.0297698974609375,
-0.039947509765625,
0.0821533203125,
0.0311431884765625,
-0.0677490234375,
0.0135040283203125,
-0.04461669921875,
-0.0070343017578125,
-0.016845703125,
-0.0005855560302734375,
-0.039947509765625,
-0.00650787353515625,
0.0181884765625,
0.04522705078125,
-0.00557708740234375,
-0.003047943115234375,
-0.01739501953125,
-0.03302001953125,
0.006000518798828125,
-0.007678985595703125,
0.0958251953125,
0.0271759033203125,
-0.0679931640625,
0.004207611083984375,
-0.06842041015625,
0.0177001953125,
0.0170745849609375,
-0.0221710205078125,
-0.01261138916015625,
-0.0148162841796875,
0.00510406494140625,
0.04144287109375,
0.045196533203125,
-0.044708251953125,
0.0230865478515625,
-0.0302734375,
0.04541015625,
0.043975830078125,
-0.0231475830078125,
0.01629638671875,
-0.0172271728515625,
0.030548095703125,
0.0316162109375,
0.0302734375,
0.02008056640625,
-0.04583740234375,
-0.05712890625,
-0.044952392578125,
0.0278472900390625,
0.053985595703125,
-0.046539306640625,
0.055389404296875,
-0.0084228515625,
-0.047027587890625,
-0.03997802734375,
0.0192718505859375,
0.043426513671875,
0.021759033203125,
0.037567138671875,
-0.0034160614013671875,
-0.04168701171875,
-0.0850830078125,
0.006076812744140625,
0.0016336441040039062,
-0.00026798248291015625,
0.01354217529296875,
0.05364990234375,
-0.0247802734375,
0.06414794921875,
-0.036346435546875,
-0.035552978515625,
-0.01297760009765625,
0.003391265869140625,
0.034515380859375,
0.057159423828125,
0.07763671875,
-0.056854248046875,
-0.04901123046875,
-0.01561737060546875,
-0.04937744140625,
0.0174560546875,
0.00022351741790771484,
-0.0201873779296875,
0.04449462890625,
0.0194854736328125,
-0.046844482421875,
0.0367431640625,
0.054290771484375,
-0.0367431640625,
0.018646240234375,
-0.0249176025390625,
0.014862060546875,
-0.07794189453125,
0.0180511474609375,
-0.002910614013671875,
-0.0205230712890625,
-0.041015625,
-0.00617218017578125,
0.01194000244140625,
0.0227813720703125,
-0.0252227783203125,
0.0262908935546875,
-0.047454833984375,
0.00688934326171875,
-0.0165557861328125,
0.017547607421875,
0.0088348388671875,
0.06256103515625,
-0.00518798828125,
0.050384521484375,
0.042083740234375,
-0.035186767578125,
0.020233154296875,
0.0408935546875,
-0.0244140625,
0.032623291015625,
-0.06494140625,
0.01416778564453125,
-0.014404296875,
0.01453399658203125,
-0.0810546875,
0.0093994140625,
0.026275634765625,
-0.038177490234375,
0.04449462890625,
-0.0101470947265625,
-0.040191650390625,
-0.040313720703125,
-0.027252197265625,
-0.00010818243026733398,
0.05615234375,
-0.0513916015625,
0.0187835693359375,
0.02825927734375,
0.00948333740234375,
-0.0540771484375,
-0.060333251953125,
-0.00882720947265625,
-0.01555633544921875,
-0.06341552734375,
0.05364990234375,
-0.0157928466796875,
-0.00823211669921875,
-0.00597381591796875,
-0.004802703857421875,
-0.01549530029296875,
0.023345947265625,
0.0270843505859375,
0.034454345703125,
-0.005641937255859375,
-0.003162384033203125,
0.005947113037109375,
0.00270843505859375,
-0.01117706298828125,
0.0011444091796875,
0.039581298828125,
-0.023345947265625,
-0.003932952880859375,
-0.0298919677734375,
0.0193328857421875,
0.04730224609375,
-0.0290679931640625,
0.060821533203125,
0.0701904296875,
-0.0213623046875,
0.002910614013671875,
-0.039398193359375,
-0.01503753662109375,
-0.03466796875,
0.0192718505859375,
-0.03302001953125,
-0.0596923828125,
0.050048828125,
0.016326904296875,
0.0212860107421875,
0.048828125,
0.045074462890625,
-0.00846099853515625,
0.09100341796875,
0.048828125,
-0.0262908935546875,
0.043426513671875,
-0.056732177734375,
-0.0024776458740234375,
-0.0755615234375,
-0.016815185546875,
-0.03375244140625,
-0.0513916015625,
-0.038848876953125,
-0.0179290771484375,
0.01494598388671875,
0.034149169921875,
-0.0251007080078125,
0.0584716796875,
-0.08038330078125,
0.0012874603271484375,
0.055633544921875,
0.0382080078125,
-0.0011816024780273438,
0.0057525634765625,
0.01169586181640625,
-0.00782012939453125,
-0.055877685546875,
-0.0301055908203125,
0.05865478515625,
0.02911376953125,
0.038055419921875,
0.0143280029296875,
0.064208984375,
0.0110015869140625,
-0.006214141845703125,
-0.0245208740234375,
0.0316162109375,
-0.0113067626953125,
-0.04266357421875,
-0.014617919921875,
-0.02728271484375,
-0.08551025390625,
0.0162353515625,
-0.01102447509765625,
-0.086669921875,
0.0312347412109375,
0.0306854248046875,
-0.035491943359375,
0.01361846923828125,
-0.039947509765625,
0.0699462890625,
-0.0110931396484375,
-0.028564453125,
-0.02203369140625,
-0.050262451171875,
0.0177459716796875,
0.0173187255859375,
-0.0125885009765625,
-0.02191162109375,
0.004848480224609375,
0.06195068359375,
-0.02301025390625,
0.058807373046875,
-0.0286407470703125,
-0.023040771484375,
0.028289794921875,
-0.003063201904296875,
0.054595947265625,
-0.003387451171875,
-0.0024013519287109375,
0.017852783203125,
0.022705078125,
-0.03314208984375,
-0.035858154296875,
0.061859130859375,
-0.0640869140625,
-0.025726318359375,
-0.033050537109375,
-0.0433349609375,
-0.01922607421875,
-0.0012378692626953125,
0.025299072265625,
0.033416748046875,
0.004093170166015625,
0.0168304443359375,
0.0633544921875,
-0.01004791259765625,
0.0367431640625,
0.038543701171875,
0.0126495361328125,
-0.0119781494140625,
0.062103271484375,
0.0086212158203125,
0.00531768798828125,
0.03668212890625,
-0.022705078125,
-0.0234832763671875,
-0.039306640625,
-0.038970947265625,
0.0068817138671875,
-0.0399169921875,
-0.0301666259765625,
-0.052520751953125,
-0.006927490234375,
-0.0262908935546875,
0.005523681640625,
-0.029632568359375,
-0.042633056640625,
-0.054290771484375,
0.02142333984375,
0.049957275390625,
0.0404052734375,
-0.0024089813232421875,
0.0099029541015625,
-0.0679931640625,
0.0118560791015625,
0.0078277587890625,
0.017425537109375,
0.0007109642028808594,
-0.041717529296875,
-0.01910400390625,
0.0255126953125,
-0.044403076171875,
-0.060546875,
0.03253173828125,
0.0016870498657226562,
0.045257568359375,
0.0016508102416992188,
0.006771087646484375,
0.048126220703125,
-0.0306396484375,
0.058929443359375,
0.0262298583984375,
-0.06121826171875,
0.052886962890625,
-0.0190277099609375,
0.021209716796875,
0.044647216796875,
0.03424072265625,
-0.001678466796875,
-0.02215576171875,
-0.059906005859375,
-0.056060791015625,
0.07379150390625,
0.03790283203125,
-0.00970458984375,
0.0084991455078125,
0.01209259033203125,
-0.01580810546875,
0.0163726806640625,
-0.02923583984375,
-0.033905029296875,
-0.01239013671875,
-0.02191162109375,
-0.004520416259765625,
-0.023651123046875,
-0.0075836181640625,
-0.035736083984375,
0.06817626953125,
-0.0019550323486328125,
0.040985107421875,
0.0364990234375,
-0.0196075439453125,
-0.0008535385131835938,
-0.00039839744567871094,
0.0638427734375,
0.06378173828125,
-0.0310516357421875,
-0.0172271728515625,
0.016204833984375,
-0.0312347412109375,
-0.00154876708984375,
0.01580810546875,
0.0015192031860351562,
0.0144500732421875,
0.0187835693359375,
0.07196044921875,
0.004253387451171875,
-0.03924560546875,
0.0282135009765625,
0.005619049072265625,
-0.032470703125,
-0.0174560546875,
-0.0006060600280761719,
-0.001277923583984375,
0.044036865234375,
0.020965576171875,
0.0111541748046875,
0.01275634765625,
-0.02850341796875,
0.01340484619140625,
0.048736572265625,
-0.041961669921875,
-0.0218963623046875,
0.0506591796875,
0.00795745849609375,
0.0015316009521484375,
0.03955078125,
-0.017852783203125,
-0.050689697265625,
0.06341552734375,
0.026123046875,
0.059661865234375,
-0.01171875,
0.004192352294921875,
0.052703857421875,
0.0250244140625,
0.00787353515625,
0.043182373046875,
0.005290985107421875,
-0.024749755859375,
-0.0218353271484375,
-0.050750732421875,
-0.00019669532775878906,
0.0224609375,
-0.0496826171875,
0.0038299560546875,
-0.0107574462890625,
-0.026092529296875,
0.01386260986328125,
0.0284423828125,
-0.0648193359375,
0.0139617919921875,
0.0091705322265625,
0.07275390625,
-0.039703369140625,
0.06329345703125,
0.0540771484375,
-0.03253173828125,
-0.049041748046875,
-0.0215606689453125,
-0.01047515869140625,
-0.061920166015625,
0.0794677734375,
0.013946533203125,
0.00748443603515625,
0.0002276897430419922,
-0.0286712646484375,
-0.07257080078125,
0.0931396484375,
0.025299072265625,
-0.069580078125,
-0.004119873046875,
-0.0008063316345214844,
0.0335693359375,
-0.0178070068359375,
0.018646240234375,
0.041107177734375,
0.036376953125,
-0.004497528076171875,
-0.0833740234375,
0.0279388427734375,
-0.0246124267578125,
0.0074005126953125,
0.01739501953125,
-0.069091796875,
0.0797119140625,
-0.0105438232421875,
0.01494598388671875,
0.01201629638671875,
0.04443359375,
0.0196075439453125,
0.006168365478515625,
0.04449462890625,
0.050384521484375,
0.044189453125,
-0.0140533447265625,
0.06683349609375,
-0.0369873046875,
0.047637939453125,
0.06817626953125,
0.0121917724609375,
0.05206298828125,
0.03497314453125,
-0.03424072265625,
0.0323486328125,
0.04998779296875,
-0.014617919921875,
0.03289794921875,
0.01280975341796875,
0.0060272216796875,
-0.0182952880859375,
0.025726318359375,
-0.035980224609375,
0.033233642578125,
0.00917816162109375,
-0.038482666015625,
-0.0140838623046875,
0.0063934326171875,
0.00518798828125,
-0.01200103759765625,
-0.0206756591796875,
0.047760009765625,
-0.0026226043701171875,
-0.05377197265625,
0.08404541015625,
-0.01617431640625,
0.0626220703125,
-0.03790283203125,
-0.01006317138671875,
-0.0031948089599609375,
0.039276123046875,
-0.02581787109375,
-0.0550537109375,
0.0175628662109375,
-0.00730133056640625,
-0.02508544921875,
-0.0067901611328125,
0.04888916015625,
-0.0283660888671875,
-0.031280517578125,
0.0279693603515625,
0.0262603759765625,
0.01143646240234375,
-0.0240478515625,
-0.089599609375,
0.02777099609375,
0.0191650390625,
-0.038818359375,
0.03778076171875,
0.01043701171875,
0.01282501220703125,
0.0360107421875,
0.0159454345703125,
-0.030670166015625,
0.0005650520324707031,
-0.018310546875,
0.07330322265625,
-0.0225372314453125,
-0.0199737548828125,
-0.0643310546875,
0.04718017578125,
-0.0168914794921875,
-0.0294647216796875,
0.06884765625,
0.03570556640625,
0.03729248046875,
-0.020050048828125,
0.038604736328125,
-0.0313720703125,
0.02508544921875,
-0.03570556640625,
0.058319091796875,
-0.06988525390625,
-0.01007843017578125,
-0.03472900390625,
-0.06982421875,
0.004001617431640625,
0.053375244140625,
-0.0020599365234375,
0.0092010498046875,
0.016021728515625,
0.049835205078125,
-0.00897979736328125,
-0.0176544189453125,
0.011474609375,
0.013519287109375,
0.0170745849609375,
0.07586669921875,
0.036651611328125,
-0.059967041015625,
0.0361328125,
-0.037200927734375,
-0.032257080078125,
-0.030029296875,
-0.057342529296875,
-0.0826416015625,
-0.05584716796875,
-0.05218505859375,
-0.039093017578125,
-0.003849029541015625,
0.0673828125,
0.07025146484375,
-0.06378173828125,
0.01534271240234375,
-0.01317596435546875,
-0.008270263671875,
-0.037628173828125,
-0.0169219970703125,
0.04046630859375,
-0.0321044921875,
-0.07794189453125,
0.022613525390625,
-0.0092926025390625,
0.023651123046875,
-0.00989532470703125,
-0.01776123046875,
-0.023956298828125,
-0.004550933837890625,
0.05841064453125,
0.0178680419921875,
-0.0487060546875,
-0.01438140869140625,
0.006134033203125,
-0.011505126953125,
0.006526947021484375,
0.00946807861328125,
-0.053466796875,
0.003826141357421875,
0.042633056640625,
0.0167236328125,
0.04345703125,
-0.0174102783203125,
0.012237548828125,
-0.05657958984375,
0.031524658203125,
0.0161590576171875,
0.0313720703125,
0.00411224365234375,
-0.035736083984375,
0.045135498046875,
-0.0105743408203125,
-0.045745849609375,
-0.06689453125,
0.007625579833984375,
-0.10693359375,
-0.02435302734375,
0.072998046875,
-0.0236663818359375,
-0.0205078125,
0.0086669921875,
-0.0273895263671875,
0.01126861572265625,
-0.0293731689453125,
0.053375244140625,
0.038238525390625,
-0.0178985595703125,
0.00287628173828125,
-0.0380859375,
0.05548095703125,
0.040374755859375,
-0.04266357421875,
0.0017223358154296875,
0.0241851806640625,
0.01983642578125,
0.041717529296875,
0.042816162109375,
-0.0021152496337890625,
0.0277252197265625,
-0.010467529296875,
0.000003039836883544922,
-0.026214599609375,
-0.016998291015625,
-0.01210784912109375,
-0.014801025390625,
-0.00879669189453125,
-0.043914794921875
]
] |
PY007/TinyLlama-1.1B-intermediate-step-240k-503b | 2023-09-19T05:36:46.000Z | [
"transformers",
"pytorch",
"safetensors",
"llama",
"text-generation",
"en",
"dataset:cerebras/SlimPajama-627B",
"dataset:bigcode/starcoderdata",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | PY007 | null | null | PY007/TinyLlama-1.1B-intermediate-step-240k-503b | 15 | 7,301 | transformers | 2023-09-16T03:19:04 | ---
license: apache-2.0
datasets:
- cerebras/SlimPajama-627B
- bigcode/starcoderdata
language:
- en
---
<div align="center">
# TinyLlama-1.1B
</div>
https://github.com/jzhang38/TinyLlama
The TinyLlama project aims to **pretrain** a **1.1B Llama model on 3 trillion tokens**. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01.
<div align="center">
<img src="./TinyLlama_logo.png" width="300"/>
</div>
We adopted exactly the same architecture and tokenizer as Llama 2. This means TinyLlama can be plugged and played in many open-source projects built upon Llama. Besides, TinyLlama is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.
#### This Model
This is an intermediate checkpoint with 240K steps and 503B tokens. **We suggest you not use this directly for inference.** The [chat model](https://huggingface.co/PY007/TinyLlama-1.1B-Chat-v0.1) is always preferred **
#### How to use
You will need the transformers>=4.31
Do check the [TinyLlama](https://github.com/jzhang38/TinyLlama) github page for more information.
```
from transformers import AutoTokenizer
import transformers
import torch
model = "PY007/TinyLlama-1.1B-intermediate-step-240k-503b"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
sequences = pipeline(
'The TinyLlama project aims to pretrain a 1.1B Llama model on 3 trillion tokens. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01.',
do_sample=True,
top_k=10,
num_return_sequences=1,
repetition_penalty=1.5,
eos_token_id=tokenizer.eos_token_id,
max_length=500,
)
for seq in sequences:
print(f"Result: {seq['generated_text']}")
``` | 2,037 | [
[
-0.017364501953125,
-0.0576171875,
0.036895751953125,
0.0242462158203125,
-0.043975830078125,
-0.006198883056640625,
-0.0172271728515625,
-0.0299072265625,
0.02569580078125,
0.0115509033203125,
-0.055419921875,
-0.02557373046875,
-0.0377197265625,
-0.0145721435546875,
-0.0186920166015625,
0.08575439453125,
0.00882720947265625,
-0.01056671142578125,
0.0201873779296875,
0.011749267578125,
-0.024505615234375,
-0.00844573974609375,
-0.07666015625,
-0.0191497802734375,
0.030120849609375,
0.04925537109375,
0.03643798828125,
0.054656982421875,
0.025787353515625,
0.0218353271484375,
-0.01425933837890625,
0.0016689300537109375,
-0.046234130859375,
-0.032745361328125,
0.0233306884765625,
-0.052886962890625,
-0.050689697265625,
0.0133514404296875,
0.041656494140625,
0.017913818359375,
-0.01178741455078125,
0.050506591796875,
-0.0019207000732421875,
0.0223846435546875,
-0.026336669921875,
0.0159912109375,
-0.04840087890625,
0.0099029541015625,
-0.0279083251953125,
0.01061248779296875,
-0.0229034423828125,
-0.024658203125,
0.0068206787109375,
-0.06488037109375,
0.010284423828125,
0.0269012451171875,
0.07452392578125,
0.0286102294921875,
-0.0129241943359375,
-0.0185089111328125,
-0.031036376953125,
0.058349609375,
-0.057037353515625,
-0.0035648345947265625,
0.027801513671875,
0.0286712646484375,
-0.0008077621459960938,
-0.08624267578125,
-0.05328369140625,
-0.0213470458984375,
0.0087432861328125,
-0.00728607177734375,
-0.01340484619140625,
-0.007045745849609375,
0.029541015625,
0.031036376953125,
-0.040618896484375,
0.0274200439453125,
-0.0472412109375,
-0.01131439208984375,
0.03765869140625,
0.03564453125,
0.005939483642578125,
-0.0200347900390625,
-0.03289794921875,
-0.01715087890625,
-0.061065673828125,
0.0020809173583984375,
0.0182647705078125,
0.02923583984375,
-0.045654296875,
0.03338623046875,
-0.00438690185546875,
0.0199737548828125,
0.01161956787109375,
-0.0236358642578125,
0.003246307373046875,
-0.030242919921875,
-0.04290771484375,
-0.00022745132446289062,
0.069091796875,
0.00017213821411132812,
-0.003711700439453125,
0.01056671142578125,
-0.0001900196075439453,
0.01300048828125,
0.00839996337890625,
-0.0811767578125,
-0.02789306640625,
0.017364501953125,
-0.03863525390625,
-0.039581298828125,
-0.024444580078125,
-0.05023193359375,
-0.006381988525390625,
0.0020847320556640625,
0.049285888671875,
-0.017425537109375,
-0.0013370513916015625,
0.0013208389282226562,
0.0206756591796875,
0.0086212158203125,
0.0259857177734375,
-0.077392578125,
0.0148162841796875,
0.04254150390625,
0.0826416015625,
0.0182342529296875,
-0.027191162109375,
-0.0204010009765625,
-0.0098419189453125,
-0.0157928466796875,
0.046295166015625,
-0.001850128173828125,
-0.0269775390625,
-0.0250396728515625,
-0.01071929931640625,
-0.007114410400390625,
-0.030120849609375,
0.014434814453125,
-0.032623291015625,
0.0175933837890625,
0.004886627197265625,
-0.029937744140625,
-0.00534820556640625,
0.0075225830078125,
-0.03802490234375,
0.07666015625,
-0.01226806640625,
-0.041351318359375,
0.0192108154296875,
-0.052154541015625,
-0.0098419189453125,
-0.008636474609375,
-0.0004773139953613281,
-0.03253173828125,
0.007419586181640625,
0.0147857666015625,
0.02154541015625,
-0.036468505859375,
0.0012493133544921875,
-0.0221710205078125,
-0.05224609375,
0.005035400390625,
-0.0105133056640625,
0.06396484375,
0.0291748046875,
-0.034637451171875,
0.01018524169921875,
-0.05126953125,
-0.004238128662109375,
0.025634765625,
-0.026123046875,
0.0122833251953125,
-0.018218994140625,
0.01364898681640625,
0.00856781005859375,
0.034759521484375,
-0.03778076171875,
0.04278564453125,
-0.048309326171875,
0.043792724609375,
0.06927490234375,
-0.01885986328125,
0.043121337890625,
-0.0233154296875,
0.037261962890625,
-0.002529144287109375,
0.0305633544921875,
-0.0096435546875,
-0.047332763671875,
-0.09991455078125,
-0.0286407470703125,
0.030731201171875,
0.0199432373046875,
-0.039886474609375,
0.0264892578125,
-0.02264404296875,
-0.059478759765625,
-0.039154052734375,
0.0142974853515625,
0.020111083984375,
0.023712158203125,
0.0263671875,
-0.0229339599609375,
-0.061492919921875,
-0.052154541015625,
0.0196990966796875,
-0.042266845703125,
-0.005916595458984375,
0.004772186279296875,
0.062744140625,
-0.0307464599609375,
0.061737060546875,
-0.03656005859375,
-0.037628173828125,
-0.01971435546875,
0.00717926025390625,
0.0290985107421875,
0.044189453125,
0.041961669921875,
-0.019989013671875,
-0.031524658203125,
-0.01380157470703125,
-0.04534912109375,
0.00848388671875,
-0.007427215576171875,
-0.007358551025390625,
-0.00739288330078125,
0.0242919921875,
-0.055389404296875,
0.03277587890625,
0.03900146484375,
-0.015594482421875,
0.019256591796875,
-0.003910064697265625,
-0.0261993408203125,
-0.07904052734375,
0.004749298095703125,
-0.01351165771484375,
-0.016845703125,
-0.038238525390625,
0.00960540771484375,
-0.00635528564453125,
-0.0173187255859375,
-0.044097900390625,
0.051483154296875,
-0.01116943359375,
0.00750732421875,
-0.033355712890625,
0.0021610260009765625,
-0.016448974609375,
0.041259765625,
-0.01016998291015625,
0.054962158203125,
0.0308837890625,
-0.038330078125,
0.0143280029296875,
0.021148681640625,
-0.024017333984375,
-0.0026035308837890625,
-0.06298828125,
0.0277557373046875,
0.0218963623046875,
0.031890869140625,
-0.06829833984375,
-0.01096343994140625,
0.047882080078125,
-0.0265655517578125,
0.015380859375,
-0.0018930435180664062,
-0.050384521484375,
-0.040557861328125,
-0.032958984375,
0.034149169921875,
0.057342529296875,
-0.05462646484375,
0.025299072265625,
0.031585693359375,
0.006870269775390625,
-0.0236358642578125,
-0.05316162109375,
0.0017108917236328125,
-0.0261993408203125,
-0.043914794921875,
0.014007568359375,
-0.00649261474609375,
0.003498077392578125,
-0.0195465087890625,
-0.0024547576904296875,
0.00958251953125,
0.01094818115234375,
0.035186767578125,
0.02386474609375,
-0.01094818115234375,
0.0016508102416992188,
-0.006114959716796875,
-0.0233306884765625,
-0.01329803466796875,
-0.0244140625,
0.052398681640625,
-0.0372314453125,
-0.016632080078125,
-0.062164306640625,
-0.01020050048828125,
0.0132904052734375,
0.021697998046875,
0.052947998046875,
0.0462646484375,
-0.03753662109375,
-0.0013399124145507812,
-0.044281005859375,
-0.022430419921875,
-0.041168212890625,
0.0089569091796875,
-0.0170440673828125,
-0.06689453125,
0.0357666015625,
0.00107574462890625,
0.00635528564453125,
0.050506591796875,
0.06793212890625,
-0.01206207275390625,
0.054962158203125,
0.05438232421875,
-0.01009368896484375,
0.039764404296875,
-0.057342529296875,
0.005725860595703125,
-0.0577392578125,
-0.0144195556640625,
-0.028045654296875,
-0.0255126953125,
-0.0277099609375,
-0.0469970703125,
0.0144500732421875,
0.013153076171875,
-0.040496826171875,
0.037445068359375,
-0.0284881591796875,
0.023712158203125,
0.032958984375,
0.00623321533203125,
0.0113525390625,
0.0024127960205078125,
-0.005268096923828125,
-0.0080413818359375,
-0.0714111328125,
-0.06805419921875,
0.10113525390625,
0.040557861328125,
0.0592041015625,
-0.008941650390625,
0.0640869140625,
-0.0003256797790527344,
0.04534912109375,
-0.046661376953125,
0.053192138671875,
0.00952911376953125,
-0.053741455078125,
-0.0073699951171875,
-0.01355743408203125,
-0.062286376953125,
0.021392822265625,
-0.01226806640625,
-0.0638427734375,
-0.0008707046508789062,
0.020111083984375,
-0.04901123046875,
0.0160064697265625,
-0.046905517578125,
0.0699462890625,
-0.0121002197265625,
-0.01287078857421875,
-0.0238494873046875,
-0.043548583984375,
0.03558349609375,
-0.0258941650390625,
0.005100250244140625,
-0.0244293212890625,
-0.0123138427734375,
0.072021484375,
-0.054046630859375,
0.07611083984375,
-0.01483917236328125,
0.006732940673828125,
0.02911376953125,
-0.0167388916015625,
0.0299072265625,
0.0190277099609375,
-0.0019121170043945312,
0.03179931640625,
-0.0085906982421875,
-0.030914306640625,
-0.00408172607421875,
0.04864501953125,
-0.0750732421875,
-0.039764404296875,
-0.039825439453125,
-0.0216827392578125,
0.00720977783203125,
0.005222320556640625,
0.0312347412109375,
-0.007843017578125,
-0.014678955078125,
0.0029888153076171875,
0.0199432373046875,
-0.004199981689453125,
0.044921875,
0.021575927734375,
-0.025726318359375,
-0.0177459716796875,
0.06158447265625,
0.0026683807373046875,
-0.003559112548828125,
0.0009670257568359375,
0.0111236572265625,
-0.01407623291015625,
-0.041748046875,
-0.047637939453125,
0.0267333984375,
-0.0322265625,
-0.022552490234375,
-0.034027099609375,
-0.0132598876953125,
-0.023406982421875,
0.003879547119140625,
-0.05474853515625,
-0.040924072265625,
-0.060150146484375,
0.00537109375,
0.01947021484375,
0.046051025390625,
-0.01398468017578125,
0.064208984375,
-0.038055419921875,
0.0183258056640625,
0.036346435546875,
-0.0014276504516601562,
0.021148681640625,
-0.07269287109375,
-0.0411376953125,
0.00658416748046875,
-0.043487548828125,
-0.04449462890625,
0.0280914306640625,
0.0201568603515625,
0.0209808349609375,
0.04754638671875,
-0.0249786376953125,
0.0855712890625,
-0.0391845703125,
0.06439208984375,
0.0244598388671875,
-0.06768798828125,
0.060455322265625,
-0.003932952880859375,
0.00957489013671875,
0.0391845703125,
0.014434814453125,
-0.01477813720703125,
-0.032745361328125,
-0.059661865234375,
-0.051422119140625,
0.07183837890625,
0.02923583984375,
0.0170135498046875,
0.00724029541015625,
0.03369140625,
-0.002658843994140625,
0.007709503173828125,
-0.064697265625,
-0.0207061767578125,
-0.0221405029296875,
-0.00860595703125,
-0.015869140625,
-0.025390625,
-0.0157470703125,
-0.035064697265625,
0.059661865234375,
-0.0120086669921875,
0.034881591796875,
-0.0152587890625,
-0.01727294921875,
-0.006855010986328125,
-0.0133514404296875,
0.05157470703125,
0.03643798828125,
-0.0164794921875,
-0.0092315673828125,
0.039093017578125,
-0.04791259765625,
0.02197265625,
0.00983428955078125,
-0.0080718994140625,
-0.00998687744140625,
0.022247314453125,
0.07275390625,
0.02777099609375,
-0.036651611328125,
0.033447265625,
-0.01812744140625,
-0.0021457672119140625,
-0.028839111328125,
0.01039886474609375,
0.0193023681640625,
0.034027099609375,
0.037994384765625,
-0.0101318359375,
-0.0088653564453125,
-0.0272064208984375,
-0.01319122314453125,
0.0177459716796875,
0.005397796630859375,
-0.033172607421875,
0.0853271484375,
0.00908660888671875,
-0.01214599609375,
0.04345703125,
-0.01175689697265625,
-0.01959228515625,
0.06524658203125,
0.036834716796875,
0.050811767578125,
0.00768280029296875,
-0.003662109375,
0.037445068359375,
0.041595458984375,
-0.0231475830078125,
0.006755828857421875,
-0.007720947265625,
-0.0313720703125,
-0.00853729248046875,
-0.06256103515625,
-0.033203125,
0.0079803466796875,
-0.0296478271484375,
0.0297393798828125,
-0.055328369140625,
-0.0218505859375,
-0.0029296875,
0.037017822265625,
-0.06561279296875,
0.0169677734375,
0.0217437744140625,
0.07501220703125,
-0.060272216796875,
0.08282470703125,
0.0465087890625,
-0.028656005859375,
-0.07958984375,
-0.016357421875,
0.008026123046875,
-0.08758544921875,
0.05584716796875,
0.032135009765625,
0.00959014892578125,
0.01082611083984375,
-0.047332763671875,
-0.05816650390625,
0.10247802734375,
0.033111572265625,
-0.0474853515625,
-0.0138092041015625,
0.0014944076538085938,
0.043365478515625,
-0.031219482421875,
0.0239410400390625,
0.043792724609375,
0.021575927734375,
0.00408172607421875,
-0.075927734375,
0.0058135986328125,
-0.021575927734375,
0.030670166015625,
-0.016693115234375,
-0.07354736328125,
0.08355712890625,
-0.02789306640625,
-0.0220184326171875,
0.048431396484375,
0.071533203125,
0.030364990234375,
0.0223541259765625,
0.03839111328125,
0.063720703125,
0.050079345703125,
-0.0208892822265625,
0.0711669921875,
-0.02716064453125,
0.053955078125,
0.06524658203125,
0.0125274658203125,
0.058013916015625,
0.04376220703125,
-0.020782470703125,
0.040069580078125,
0.0809326171875,
-0.00592803955078125,
0.0418701171875,
0.00920867919921875,
-0.0101470947265625,
-0.017822265625,
0.0095062255859375,
-0.047210693359375,
0.033599853515625,
0.029083251953125,
-0.0190887451171875,
-0.002079010009765625,
0.0073089599609375,
0.001735687255859375,
-0.04278564453125,
-0.02423095703125,
0.042205810546875,
0.0236358642578125,
-0.0140228271484375,
0.047393798828125,
0.0170440673828125,
0.07037353515625,
-0.04803466796875,
0.0218353271484375,
-0.033935546875,
0.01284027099609375,
-0.0038928985595703125,
-0.004344940185546875,
-0.0004506111145019531,
0.0162353515625,
0.01561737060546875,
-0.009674072265625,
0.04913330078125,
-0.006549835205078125,
-0.043670654296875,
-0.0018062591552734375,
0.0223388671875,
0.0194091796875,
0.008941650390625,
-0.049072265625,
0.027313232421875,
-0.009735107421875,
-0.04315185546875,
0.0267181396484375,
0.003753662109375,
0.0193023681640625,
0.049468994140625,
0.052337646484375,
0.00691986083984375,
0.0230560302734375,
-0.0211944580078125,
0.07269287109375,
-0.03515625,
-0.04931640625,
-0.07672119140625,
0.021331787109375,
0.00666046142578125,
-0.0372314453125,
0.06243896484375,
0.048004150390625,
0.0555419921875,
-0.008026123046875,
0.022979736328125,
-0.01239776611328125,
0.007965087890625,
-0.033050537109375,
0.049774169921875,
-0.06024169921875,
0.0239410400390625,
-0.0036487579345703125,
-0.056182861328125,
-0.006519317626953125,
0.07952880859375,
-0.0095367431640625,
-0.0009036064147949219,
0.034881591796875,
0.06317138671875,
-0.004772186279296875,
0.01690673828125,
-0.005725860595703125,
0.0150299072265625,
0.0231170654296875,
0.060394287109375,
0.070068359375,
-0.063720703125,
0.05523681640625,
-0.03741455078125,
-0.01392364501953125,
-0.03753662109375,
-0.05010986328125,
-0.0640869140625,
-0.019317626953125,
-0.0215606689453125,
-0.01433563232421875,
-0.005306243896484375,
0.0771484375,
0.06365966796875,
-0.0435791015625,
-0.02618408203125,
0.003498077392578125,
0.0003173351287841797,
0.00266265869140625,
-0.0096893310546875,
0.02197265625,
-0.0166778564453125,
-0.07086181640625,
0.0251007080078125,
0.015411376953125,
0.0212249755859375,
-0.027496337890625,
-0.01348876953125,
-0.01175689697265625,
-0.0007386207580566406,
0.0310211181640625,
0.03472900390625,
-0.053680419921875,
-0.03057861328125,
-0.01898193359375,
-0.032257080078125,
0.001178741455078125,
0.04437255859375,
-0.056243896484375,
0.0204620361328125,
0.006072998046875,
0.0272674560546875,
0.07452392578125,
-0.0299835205078125,
0.002681732177734375,
-0.048065185546875,
0.054718017578125,
0.005809783935546875,
0.026336669921875,
0.007785797119140625,
-0.003696441650390625,
0.048797607421875,
0.0172882080078125,
-0.043365478515625,
-0.07391357421875,
-0.0067596435546875,
-0.07330322265625,
0.007595062255859375,
0.07220458984375,
-0.005001068115234375,
-0.0135955810546875,
0.019989013671875,
-0.0222320556640625,
0.0352783203125,
-0.007297515869140625,
0.07647705078125,
0.02777099609375,
-0.0015125274658203125,
0.00954437255859375,
-0.037567138671875,
0.027191162109375,
0.035491943359375,
-0.06134033203125,
-0.034759521484375,
0.0124969482421875,
0.02960205078125,
0.010833740234375,
0.08984375,
0.021575927734375,
0.029510498046875,
0.0177764892578125,
-0.00402069091796875,
-0.00839996337890625,
-0.0223388671875,
-0.025177001953125,
0.00191497802734375,
0.00262451171875,
-0.0294036865234375
]
] |
hustvl/vitmatte-small-composition-1k | 2023-09-21T09:25:26.000Z | [
"transformers",
"pytorch",
"vitmatte",
"vision",
"arxiv:2305.15272",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | null | hustvl | null | null | hustvl/vitmatte-small-composition-1k | 1 | 7,295 | transformers | 2023-09-10T07:47:27 | ---
license: apache-2.0
tags:
- vision
---
# ViTMatte model
ViTMatte model trained on Composition-1k. It was introduced in the paper [ViTMatte: Boosting Image Matting with Pretrained Plain Vision Transformers](https://arxiv.org/abs/2305.15272) by Yao et al. and first released in [this repository](https://github.com/hustvl/ViTMatte).
Disclaimer: The team releasing ViTMatte did not write a model card for this model so this model card has been written by the Hugging Face team.
## Model description
ViTMatte is a simple approach to image matting, the task of accurately estimating the foreground object in an image. The model consists of a Vision Transformer (ViT) with a lightweight head on top.
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/vitmatte_architecture.png"
alt="drawing" width="600"/>
<small> ViTMatte high-level overview. Taken from the <a href="https://arxiv.org/abs/2305.15272">original paper.</a> </small>
## Intended uses & limitations
You can use the raw model for image matting. See the [model hub](https://huggingface.co/models?search=vitmatte) to look for other
fine-tuned versions that may interest you.
### How to use
We refer to the [docs](https://huggingface.co/docs/transformers/main/en/model_doc/vitmatte#transformers.VitMatteForImageMatting.forward.example).
### BibTeX entry and citation info
```bibtex
@misc{yao2023vitmatte,
title={ViTMatte: Boosting Image Matting with Pretrained Plain Vision Transformers},
author={Jingfeng Yao and Xinggang Wang and Shusheng Yang and Baoyuan Wang},
year={2023},
eprint={2305.15272},
archivePrefix={arXiv},
primaryClass={cs.CV}
}
``` | 1,716 | [
[
-0.054962158203125,
-0.04278564453125,
0.01027679443359375,
0.01263427734375,
-0.0287017822265625,
-0.0282440185546875,
0.002574920654296875,
-0.0291900634765625,
0.02471923828125,
0.0305938720703125,
-0.05560302734375,
-0.031036376953125,
-0.05084228515625,
-0.02459716796875,
-0.03643798828125,
0.08624267578125,
-0.01070404052734375,
0.00939178466796875,
-0.02508544921875,
-0.0321044921875,
-0.0400390625,
-0.0174407958984375,
-0.0169525146484375,
-0.0247344970703125,
0.00844573974609375,
0.02734375,
0.037139892578125,
0.0311431884765625,
0.061676025390625,
0.02178955078125,
-0.01513671875,
-0.01180267333984375,
-0.044677734375,
-0.00908660888671875,
0.005687713623046875,
-0.0347900390625,
-0.04608154296875,
0.0242156982421875,
0.04052734375,
0.00635528564453125,
-0.0087890625,
0.0399169921875,
0.01531982421875,
0.048004150390625,
-0.050384521484375,
0.005580902099609375,
-0.028045654296875,
0.0189666748046875,
-0.0255584716796875,
0.0002868175506591797,
-0.01226806640625,
-0.006168365478515625,
0.009063720703125,
-0.051605224609375,
0.027618408203125,
-0.024810791015625,
0.11151123046875,
0.027984619140625,
0.001941680908203125,
0.03765869140625,
-0.0460205078125,
0.049560546875,
-0.0188751220703125,
0.0167388916015625,
0.01776123046875,
0.0430908203125,
0.029296875,
-0.06414794921875,
-0.034088134765625,
0.0015630722045898438,
-0.0013208389282226562,
0.03607177734375,
-0.004611968994140625,
-0.001903533935546875,
0.053436279296875,
0.0404052734375,
-0.02801513671875,
-0.03814697265625,
-0.052215576171875,
0.00374603271484375,
0.0318603515625,
-0.0013494491577148438,
0.03448486328125,
-0.0095062255859375,
-0.0745849609375,
-0.0225372314453125,
-0.0276641845703125,
0.00708770751953125,
-0.007190704345703125,
-0.01031494140625,
-0.05157470703125,
0.055023193359375,
-0.01273345947265625,
0.045623779296875,
0.0084228515625,
0.00913238525390625,
0.0227813720703125,
-0.0242462158203125,
-0.02264404296875,
-0.033172607421875,
0.0450439453125,
0.06103515625,
0.05059814453125,
0.00959014892578125,
-0.0234222412109375,
0.01248931884765625,
0.042724609375,
-0.0902099609375,
-0.0260009765625,
-0.023101806640625,
-0.0263214111328125,
-0.0345458984375,
0.0192413330078125,
-0.0633544921875,
0.0016183853149414062,
-0.0340576171875,
0.040863037109375,
-0.0290069580078125,
-0.018768310546875,
-0.013397216796875,
-0.00292205810546875,
0.058349609375,
0.0401611328125,
-0.0443115234375,
0.0230255126953125,
0.03076171875,
0.08062744140625,
-0.0005016326904296875,
-0.01261138916015625,
-0.0211181640625,
-0.016082763671875,
-0.01256561279296875,
0.0576171875,
-0.0290985107421875,
-0.0290679931640625,
-0.00632476806640625,
0.017578125,
0.0037288665771484375,
-0.0236053466796875,
0.04339599609375,
-0.01096343994140625,
0.0138702392578125,
-0.0212554931640625,
-0.0006108283996582031,
-0.0146942138671875,
0.0174713134765625,
-0.03076171875,
0.0750732421875,
0.0297698974609375,
-0.0692138671875,
0.0426025390625,
-0.042694091796875,
0.0034961700439453125,
0.01493072509765625,
0.0023651123046875,
-0.0562744140625,
0.00690460205078125,
0.0115203857421875,
0.0254058837890625,
-0.01268768310546875,
0.0007576942443847656,
-0.0216522216796875,
-0.026153564453125,
-0.007686614990234375,
-0.005893707275390625,
0.06048583984375,
0.009429931640625,
-0.010650634765625,
0.00977325439453125,
-0.036834716796875,
-0.00664520263671875,
0.031890869140625,
0.00948333740234375,
-0.01264190673828125,
-0.0201416015625,
0.0218658447265625,
0.0196685791015625,
0.00940704345703125,
-0.040008544921875,
0.015411376953125,
-0.0034847259521484375,
0.01181793212890625,
0.04486083984375,
-0.005401611328125,
0.0279388427734375,
-0.049041748046875,
0.048065185546875,
0.01360321044921875,
0.045013427734375,
-0.030181884765625,
-0.05279541015625,
-0.07586669921875,
-0.0189666748046875,
0.0182647705078125,
0.0380859375,
-0.0408935546875,
0.0019140243530273438,
-0.0265960693359375,
-0.041900634765625,
-0.04541015625,
0.0120391845703125,
0.041107177734375,
0.045501708984375,
0.0418701171875,
-0.052825927734375,
-0.0285491943359375,
-0.07977294921875,
-0.01448822021484375,
0.014007568359375,
0.0142822265625,
0.0024967193603515625,
0.03167724609375,
-0.007289886474609375,
0.074462890625,
-0.0338134765625,
-0.0306396484375,
0.0150299072265625,
0.00023245811462402344,
0.0003657341003417969,
0.05499267578125,
0.06475830078125,
-0.06536865234375,
-0.057708740234375,
-0.0159149169921875,
-0.0694580078125,
0.005306243896484375,
0.032684326171875,
-0.031890869140625,
-0.004009246826171875,
0.0249786376953125,
-0.0548095703125,
0.055816650390625,
0.032318115234375,
-0.00511932373046875,
0.050018310546875,
0.01708984375,
0.00963592529296875,
-0.08734130859375,
0.010101318359375,
0.04901123046875,
-0.048187255859375,
-0.048065185546875,
0.0341796875,
-0.0032329559326171875,
-0.0194244384765625,
-0.0501708984375,
0.047637939453125,
-0.054107666015625,
0.0226287841796875,
-0.015838623046875,
0.00203704833984375,
-0.00548553466796875,
0.045989990234375,
0.0004737377166748047,
0.049285888671875,
0.047943115234375,
-0.029144287109375,
0.01067352294921875,
0.05780029296875,
-0.020263671875,
0.0869140625,
-0.066162109375,
0.005584716796875,
-0.025421142578125,
0.0036754608154296875,
-0.058380126953125,
-0.0152435302734375,
0.04205322265625,
-0.0372314453125,
0.04052734375,
-0.04180908203125,
-0.0081939697265625,
-0.040252685546875,
-0.0240478515625,
0.040679931640625,
0.07769775390625,
-0.031036376953125,
0.03643798828125,
0.040802001953125,
0.0035839080810546875,
-0.038482666015625,
-0.06292724609375,
-0.0087738037109375,
-0.026397705078125,
-0.06280517578125,
0.02886962890625,
-0.0157928466796875,
-0.01284027099609375,
-0.004489898681640625,
-0.0186920166015625,
-0.01251983642578125,
-0.00751495361328125,
0.030853271484375,
0.0269622802734375,
-0.025421142578125,
-0.014251708984375,
-0.02166748046875,
-0.04046630859375,
0.0005321502685546875,
0.004039764404296875,
0.018035888671875,
-0.02996826171875,
-0.01837158203125,
-0.0396728515625,
0.00528717041015625,
0.034393310546875,
-0.0189666748046875,
0.0430908203125,
0.069580078125,
-0.04400634765625,
-0.01446533203125,
-0.048095703125,
-0.0296478271484375,
-0.036041259765625,
-0.007335662841796875,
-0.03887939453125,
-0.034576416015625,
0.035675048828125,
0.016143798828125,
0.006320953369140625,
0.048004150390625,
0.043701171875,
0.0007529258728027344,
0.05841064453125,
0.05865478515625,
0.01232147216796875,
0.0643310546875,
-0.04168701171875,
-0.007274627685546875,
-0.072265625,
-0.0197601318359375,
-0.0257110595703125,
-0.021942138671875,
-0.03167724609375,
-0.047637939453125,
0.03350830078125,
0.01036834716796875,
-0.01482391357421875,
0.0462646484375,
-0.06793212890625,
0.03814697265625,
0.0357666015625,
0.0257720947265625,
0.01082611083984375,
0.0218658447265625,
-0.00370025634765625,
-0.0103607177734375,
-0.035003662109375,
-0.0282745361328125,
0.0526123046875,
0.024627685546875,
0.0703125,
-0.0130767822265625,
0.034942626953125,
-0.004940032958984375,
0.043975830078125,
-0.05633544921875,
0.043212890625,
0.0038604736328125,
-0.053436279296875,
0.007335662841796875,
-0.0144500732421875,
-0.04827880859375,
0.0189361572265625,
-0.03179931640625,
-0.061431884765625,
0.012481689453125,
0.0232391357421875,
-0.00408172607421875,
0.015228271484375,
-0.048858642578125,
0.0640869140625,
0.00592041015625,
-0.037933349609375,
0.001987457275390625,
-0.0599365234375,
0.047637939453125,
0.00847625732421875,
-0.01413726806640625,
-0.019866943359375,
0.03656005859375,
0.056671142578125,
-0.044403076171875,
0.058990478515625,
-0.033447265625,
0.01885986328125,
0.03302001953125,
0.001941680908203125,
0.024505615234375,
0.0014829635620117188,
0.01448822021484375,
0.0303192138671875,
0.012908935546875,
-0.0394287109375,
-0.0272674560546875,
0.038238525390625,
-0.05230712890625,
-0.046295166015625,
-0.033782958984375,
-0.032073974609375,
0.0009369850158691406,
0.01971435546875,
0.04498291015625,
0.034942626953125,
-0.02288818359375,
0.0203704833984375,
0.040252685546875,
0.006084442138671875,
0.021453857421875,
0.0172882080078125,
-0.0303497314453125,
-0.0197296142578125,
0.06512451171875,
0.0131988525390625,
0.0229949951171875,
0.0172882080078125,
0.0133514404296875,
0.00859832763671875,
-0.011749267578125,
-0.027008056640625,
0.02874755859375,
-0.0535888671875,
-0.019287109375,
-0.053009033203125,
-0.044677734375,
-0.00957489013671875,
-0.0179901123046875,
-0.043212890625,
-0.02960205078125,
-0.0167083740234375,
0.01345062255859375,
0.03924560546875,
0.037078857421875,
0.00591278076171875,
0.041046142578125,
-0.052001953125,
0.0279083251953125,
0.01922607421875,
0.035247802734375,
-0.021942138671875,
-0.0303192138671875,
-0.0206146240234375,
0.0240020751953125,
-0.049530029296875,
-0.05316162109375,
0.01352691650390625,
0.002841949462890625,
0.04443359375,
0.0282745361328125,
-0.0167388916015625,
0.049224853515625,
-0.021331787109375,
0.05145263671875,
0.031524658203125,
-0.06170654296875,
0.042144775390625,
-0.034759521484375,
0.018280029296875,
0.053955078125,
0.0179443359375,
-0.034881591796875,
0.00093841552734375,
-0.06292724609375,
-0.0498046875,
0.049072265625,
0.02294921875,
0.023193359375,
0.035736083984375,
0.032073974609375,
-0.0096893310546875,
0.019073486328125,
-0.07623291015625,
-0.0225067138671875,
-0.0361328125,
-0.005340576171875,
-0.0008587837219238281,
-0.0229339599609375,
-0.036285400390625,
-0.048980712890625,
0.0498046875,
-0.012237548828125,
0.051422119140625,
0.018402099609375,
-0.0224456787109375,
-0.02313232421875,
-0.03466796875,
0.050811767578125,
0.04119873046875,
-0.017303466796875,
-0.024932861328125,
0.01515960693359375,
-0.034942626953125,
-0.00371551513671875,
0.004390716552734375,
-0.023101806640625,
0.000873565673828125,
0.0064239501953125,
0.07147216796875,
0.0007786750793457031,
-0.003444671630859375,
0.06982421875,
0.0008335113525390625,
-0.0284423828125,
-0.033477783203125,
0.0083160400390625,
-0.0148773193359375,
0.0122222900390625,
-0.00537872314453125,
0.025634765625,
-0.0026760101318359375,
-0.005252838134765625,
0.025665283203125,
0.045623779296875,
-0.056549072265625,
-0.0340576171875,
0.0625,
0.00582122802734375,
-0.0240325927734375,
0.038177490234375,
0.004161834716796875,
-0.03289794921875,
0.046630859375,
0.0230255126953125,
0.07733154296875,
-0.032562255859375,
0.01519012451171875,
0.049957275390625,
0.029876708984375,
-0.00916290283203125,
0.01488494873046875,
-0.01358795166015625,
-0.046844482421875,
-0.003368377685546875,
-0.048370361328125,
-0.0190582275390625,
-0.003421783447265625,
-0.05950927734375,
0.0626220703125,
-0.036773681640625,
-0.02532958984375,
-0.0056610107421875,
-0.01171112060546875,
-0.08123779296875,
0.0214385986328125,
0.016357421875,
0.08453369140625,
-0.05401611328125,
0.041412353515625,
0.056121826171875,
-0.060699462890625,
-0.04437255859375,
0.0037746429443359375,
0.01210784912109375,
-0.07513427734375,
0.0187835693359375,
0.029205322265625,
-0.0014314651489257812,
0.0049591064453125,
-0.0692138671875,
-0.04351806640625,
0.081787109375,
0.0169830322265625,
-0.03289794921875,
-0.0051116943359375,
-0.0203399658203125,
0.0220947265625,
-0.02044677734375,
0.0224456787109375,
0.006626129150390625,
0.043914794921875,
0.031585693359375,
-0.042083740234375,
0.00955963134765625,
-0.049163818359375,
0.015777587890625,
0.00943756103515625,
-0.0787353515625,
0.0633544921875,
-0.00699615478515625,
-0.019683837890625,
0.0228118896484375,
0.04248046875,
-0.0019235610961914062,
0.015716552734375,
0.0379638671875,
0.068603515625,
0.0183868408203125,
-0.0278472900390625,
0.0908203125,
-0.036468505859375,
0.048736572265625,
0.0523681640625,
0.0031223297119140625,
0.0198822021484375,
0.0197296142578125,
-0.0020427703857421875,
0.032623291015625,
0.036346435546875,
-0.035064697265625,
0.033294677734375,
0.0073394775390625,
-0.0018377304077148438,
-0.0174713134765625,
-0.004993438720703125,
-0.04547119140625,
0.0260009765625,
0.01377105712890625,
-0.01947021484375,
-0.0275421142578125,
0.010955810546875,
0.01568603515625,
-0.0219879150390625,
-0.0032024383544921875,
0.04779052734375,
0.033935546875,
-0.04168701171875,
0.049163818359375,
-0.01474761962890625,
0.041046142578125,
-0.039154052734375,
-0.00713348388671875,
-0.011199951171875,
0.00605010986328125,
-0.01255035400390625,
-0.052215576171875,
0.0204315185546875,
-0.004695892333984375,
-0.004474639892578125,
-0.019683837890625,
0.0677490234375,
-0.017120361328125,
-0.0479736328125,
0.01134490966796875,
0.02874755859375,
0.0173492431640625,
0.00977325439453125,
-0.06951904296875,
0.0008258819580078125,
-0.0303497314453125,
-0.041473388671875,
0.0089263916015625,
0.0029811859130859375,
0.0006313323974609375,
0.04541015625,
0.0321044921875,
-0.020660400390625,
-0.011444091796875,
-0.011749267578125,
0.07037353515625,
-0.032501220703125,
-0.03460693359375,
-0.050018310546875,
0.07513427734375,
-0.0139007568359375,
-0.037322998046875,
0.062347412109375,
0.0256805419921875,
0.04937744140625,
-0.0081329345703125,
0.048004150390625,
0.00867462158203125,
0.005764007568359375,
-0.040313720703125,
0.051666259765625,
-0.060546875,
-0.02813720703125,
-0.046844482421875,
-0.09814453125,
-0.0169525146484375,
0.07281494140625,
-0.0106964111328125,
0.046295166015625,
0.041900634765625,
0.0628662109375,
-0.03790283203125,
-0.00421905517578125,
0.0207061767578125,
0.0229034423828125,
0.0133819580078125,
0.002941131591796875,
0.0528564453125,
-0.0406494140625,
0.004543304443359375,
-0.0280914306640625,
-0.023895263671875,
-0.022247314453125,
-0.06304931640625,
-0.0677490234375,
-0.060943603515625,
-0.043243408203125,
-0.057891845703125,
0.0123443603515625,
0.04052734375,
0.0975341796875,
-0.0372314453125,
0.002719879150390625,
-0.01082611083984375,
-0.009063720703125,
-0.00875091552734375,
-0.0175018310546875,
-0.007457733154296875,
0.00327301025390625,
-0.060516357421875,
-0.012054443359375,
0.00879669189453125,
0.026123046875,
-0.0257720947265625,
0.005771636962890625,
-0.0027942657470703125,
0.00861358642578125,
0.0253753662109375,
0.0216217041015625,
-0.022857666015625,
-0.022430419921875,
-0.019561767578125,
-0.007259368896484375,
0.032073974609375,
0.048187255859375,
-0.0377197265625,
0.031585693359375,
0.039276123046875,
0.0174407958984375,
0.08892822265625,
0.024078369140625,
0.0266876220703125,
-0.048492431640625,
0.026763916015625,
-0.00974273681640625,
0.037872314453125,
0.03057861328125,
-0.0087432861328125,
0.05474853515625,
0.045867919921875,
-0.0426025390625,
-0.05364990234375,
0.016021728515625,
-0.1141357421875,
0.0166015625,
0.07513427734375,
0.0011806488037109375,
-0.04052734375,
0.0306854248046875,
-0.020904541015625,
0.016632080078125,
-0.005138397216796875,
0.030426025390625,
0.044189453125,
0.03924560546875,
-0.04168701171875,
-0.057586669921875,
0.031036376953125,
0.0166015625,
-0.047149658203125,
-0.03057861328125,
0.025238037109375,
0.044525146484375,
0.0252685546875,
0.049530029296875,
-0.0284576416015625,
0.02294921875,
0.002773284912109375,
0.03973388671875,
-0.01152801513671875,
-0.03448486328125,
-0.01372528076171875,
-0.0037555694580078125,
-0.01385498046875,
-0.0206451416015625
]
] |
Helsinki-NLP/opus-mt-sk-en | 2023-08-16T12:04:00.000Z | [
"transformers",
"pytorch",
"tf",
"marian",
"text2text-generation",
"translation",
"sk",
"en",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | translation | Helsinki-NLP | null | null | Helsinki-NLP/opus-mt-sk-en | 0 | 7,292 | transformers | 2022-03-02T23:29:04 | ---
tags:
- translation
license: apache-2.0
---
### opus-mt-sk-en
* source languages: sk
* target languages: en
* OPUS readme: [sk-en](https://github.com/Helsinki-NLP/OPUS-MT-train/blob/master/models/sk-en/README.md)
* dataset: opus
* model: transformer-align
* pre-processing: normalization + SentencePiece
* download original weights: [opus-2020-01-16.zip](https://object.pouta.csc.fi/OPUS-MT-models/sk-en/opus-2020-01-16.zip)
* test set translations: [opus-2020-01-16.test.txt](https://object.pouta.csc.fi/OPUS-MT-models/sk-en/opus-2020-01-16.test.txt)
* test set scores: [opus-2020-01-16.eval.txt](https://object.pouta.csc.fi/OPUS-MT-models/sk-en/opus-2020-01-16.eval.txt)
## Benchmarks
| testset | BLEU | chr-F |
|-----------------------|-------|-------|
| JW300.sk.en | 42.2 | 0.612 |
| 816 | [
[
-0.014129638671875,
-0.0233612060546875,
0.0229949951171875,
0.032928466796875,
-0.035552978515625,
-0.0279998779296875,
-0.033477783203125,
-0.006317138671875,
-0.0012464523315429688,
0.0389404296875,
-0.05206298828125,
-0.04351806640625,
-0.040985107421875,
0.01448822021484375,
-0.0016813278198242188,
0.051177978515625,
-0.01141357421875,
0.045989990234375,
0.01149749755859375,
-0.03466796875,
-0.0224151611328125,
-0.03363037109375,
-0.02557373046875,
-0.0247650146484375,
0.0194244384765625,
0.032196044921875,
0.0196075439453125,
0.02838134765625,
0.06976318359375,
0.01727294921875,
-0.0106201171875,
-0.00658416748046875,
-0.036346435546875,
0.0018110275268554688,
0.0038013458251953125,
-0.048187255859375,
-0.056182861328125,
-0.01038360595703125,
0.07427978515625,
0.034088134765625,
0.002666473388671875,
0.0318603515625,
-0.004123687744140625,
0.0709228515625,
-0.018310546875,
0.01192474365234375,
-0.04730224609375,
0.00537872314453125,
-0.0204620361328125,
-0.0163116455078125,
-0.053375244140625,
-0.0175933837890625,
0.0170440673828125,
-0.050262451171875,
-0.005157470703125,
0.01036834716796875,
0.1007080078125,
0.0198974609375,
-0.0277099609375,
-0.014190673828125,
-0.0440673828125,
0.07806396484375,
-0.064208984375,
0.046539306640625,
0.0303802490234375,
0.0208587646484375,
0.0160369873046875,
-0.037139892578125,
-0.0208740234375,
0.0003399848937988281,
-0.0170440673828125,
0.0205078125,
0.004131317138671875,
-0.01006317138671875,
0.02294921875,
0.054473876953125,
-0.055999755859375,
-0.00308990478515625,
-0.051483154296875,
-0.00370025634765625,
0.05609130859375,
0.008270263671875,
0.00978851318359375,
-0.016937255859375,
-0.0302734375,
-0.04296875,
-0.05535888671875,
0.0103912353515625,
0.032867431640625,
0.0229339599609375,
-0.02838134765625,
0.0614013671875,
-0.0183868408203125,
0.040313720703125,
-0.002323150634765625,
-0.00403594970703125,
0.072021484375,
-0.0288848876953125,
-0.022369384765625,
-0.00748443603515625,
0.08172607421875,
0.0279998779296875,
0.00344085693359375,
0.002655029296875,
-0.015167236328125,
-0.01374053955078125,
0.0082550048828125,
-0.059661865234375,
-0.010833740234375,
0.008575439453125,
-0.035980224609375,
-0.01160430908203125,
0.0028934478759765625,
-0.0458984375,
0.019256591796875,
-0.03753662109375,
0.04949951171875,
-0.0467529296875,
-0.022125244140625,
0.02392578125,
-0.0007266998291015625,
0.0299835205078125,
-0.004161834716796875,
-0.0386962890625,
0.0082550048828125,
0.028564453125,
0.0528564453125,
-0.035400390625,
-0.015655517578125,
-0.0276336669921875,
-0.01202392578125,
-0.01207733154296875,
0.048309326171875,
-0.005138397216796875,
-0.03192138671875,
-0.00679779052734375,
0.03643798828125,
-0.0251312255859375,
-0.0253448486328125,
0.09661865234375,
-0.0167694091796875,
0.048980712890625,
-0.0323486328125,
-0.039764404296875,
-0.026275634765625,
0.03631591796875,
-0.04241943359375,
0.1031494140625,
0.0130462646484375,
-0.05987548828125,
0.0211334228515625,
-0.0609130859375,
-0.0117034912109375,
0.0031681060791015625,
0.003635406494140625,
-0.041015625,
0.014068603515625,
0.0017023086547851562,
0.033477783203125,
-0.01837158203125,
0.025299072265625,
-0.0038890838623046875,
-0.0225372314453125,
-0.0012197494506835938,
-0.0251922607421875,
0.06536865234375,
0.01727294921875,
-0.0244598388671875,
0.01467132568359375,
-0.074462890625,
0.0038738250732421875,
-0.003025054931640625,
-0.042724609375,
-0.01861572265625,
0.011688232421875,
0.0184326171875,
0.0166473388671875,
0.0213165283203125,
-0.05078125,
0.020721435546875,
-0.05706787109375,
0.01189422607421875,
0.0509033203125,
-0.0265350341796875,
0.03045654296875,
-0.0263824462890625,
0.0244598388671875,
0.00868988037109375,
0.006378173828125,
0.0008912086486816406,
-0.03253173828125,
-0.06524658203125,
-0.01482391357421875,
0.04290771484375,
0.08135986328125,
-0.060821533203125,
0.05804443359375,
-0.05206298828125,
-0.061767578125,
-0.054656982421875,
-0.0104827880859375,
0.036285400390625,
0.0219573974609375,
0.0322265625,
-0.0103607177734375,
-0.0330810546875,
-0.083984375,
-0.01218414306640625,
-0.0161895751953125,
-0.01702880859375,
0.02294921875,
0.046417236328125,
-0.00734710693359375,
0.04351806640625,
-0.038482666015625,
-0.022857666015625,
-0.01708984375,
0.0129241943359375,
0.039764404296875,
0.05108642578125,
0.03985595703125,
-0.0673828125,
-0.054473876953125,
0.0003211498260498047,
-0.048553466796875,
-0.0149078369140625,
0.01041412353515625,
-0.0166473388671875,
0.00920867919921875,
0.007457733154296875,
-0.023040771484375,
0.00908660888671875,
0.047027587890625,
-0.0406494140625,
0.0509033203125,
-0.00909423828125,
0.0208740234375,
-0.10321044921875,
0.01043701171875,
-0.01708984375,
-0.00229644775390625,
-0.0355224609375,
-0.0038909912109375,
0.0156707763671875,
0.006687164306640625,
-0.062469482421875,
0.042266845703125,
-0.01537322998046875,
-0.0080108642578125,
0.0185394287109375,
-0.0017681121826171875,
0.0010480880737304688,
0.05767822265625,
-0.0005536079406738281,
0.06005859375,
0.052764892578125,
-0.039794921875,
0.0115966796875,
0.0404052734375,
-0.036285400390625,
0.032867431640625,
-0.06085205078125,
-0.0201568603515625,
0.0254974365234375,
-0.00707244873046875,
-0.052642822265625,
0.002223968505859375,
0.025543212890625,
-0.050811767578125,
0.02789306640625,
-0.01111602783203125,
-0.056396484375,
-0.00762176513671875,
-0.0218048095703125,
0.035736083984375,
0.048126220703125,
-0.01529693603515625,
0.049041748046875,
-0.00044155120849609375,
-0.00949859619140625,
-0.029296875,
-0.07366943359375,
-0.014801025390625,
-0.0301361083984375,
-0.055572509765625,
0.01169586181640625,
-0.0341796875,
-0.0020275115966796875,
0.0013141632080078125,
0.01499176025390625,
-0.010040283203125,
0.0017528533935546875,
0.005054473876953125,
0.0200958251953125,
-0.041107177734375,
0.01079559326171875,
0.00579833984375,
-0.01499176025390625,
-0.0091094970703125,
-0.013092041015625,
0.044769287109375,
-0.0220947265625,
-0.0188751220703125,
-0.04443359375,
0.006267547607421875,
0.050445556640625,
-0.02099609375,
0.0604248046875,
0.03851318359375,
-0.006816864013671875,
0.016143798828125,
-0.03369140625,
0.006671905517578125,
-0.034027099609375,
0.01204681396484375,
-0.03228759765625,
-0.056396484375,
0.038787841796875,
0.0013103485107421875,
0.0305023193359375,
0.06488037109375,
0.04461669921875,
0.00365447998046875,
0.056671142578125,
0.02337646484375,
-0.0002789497375488281,
0.027587890625,
-0.03582763671875,
-0.005859375,
-0.08306884765625,
0.0036830902099609375,
-0.049835205078125,
-0.0246429443359375,
-0.061492919921875,
-0.0158538818359375,
0.0179595947265625,
0.00848388671875,
-0.016265869140625,
0.05328369140625,
-0.0419921875,
0.0160980224609375,
0.042510986328125,
-0.012969970703125,
0.018096923828125,
-0.00193023681640625,
-0.049285888671875,
-0.023101806640625,
-0.03472900390625,
-0.04083251953125,
0.0968017578125,
0.0225372314453125,
0.0244903564453125,
0.0100250244140625,
0.02960205078125,
-0.00044798851013183594,
0.0160980224609375,
-0.044525146484375,
0.03717041015625,
-0.0201416015625,
-0.058807373046875,
-0.020904541015625,
-0.042572021484375,
-0.0489501953125,
0.040618896484375,
-0.0204010009765625,
-0.045989990234375,
0.0109710693359375,
0.003025054931640625,
-0.004802703857421875,
0.031768798828125,
-0.05218505859375,
0.08062744140625,
-0.01043701171875,
-0.00286865234375,
0.0223388671875,
-0.034942626953125,
0.0198822021484375,
0.0013751983642578125,
0.01369476318359375,
-0.0200653076171875,
0.01399993896484375,
0.045501708984375,
-0.0010089874267578125,
0.031585693359375,
-0.0068206787109375,
-0.0023136138916015625,
0.0005097389221191406,
0.0021419525146484375,
0.0302276611328125,
-0.00797271728515625,
-0.0309600830078125,
0.029083251953125,
0.0046234130859375,
-0.0286865234375,
-0.0094757080078125,
0.037567138671875,
-0.047760009765625,
0.0016231536865234375,
-0.03436279296875,
-0.050445556640625,
0.0016326904296875,
0.025360107421875,
0.0535888671875,
0.049285888671875,
-0.017333984375,
0.048828125,
0.05804443359375,
-0.0274658203125,
0.0224151611328125,
0.058013916015625,
-0.01381683349609375,
-0.0440673828125,
0.066162109375,
0.01248931884765625,
0.035186767578125,
0.0362548828125,
0.01062774658203125,
-0.0106048583984375,
-0.0545654296875,
-0.051666259765625,
0.0191497802734375,
-0.020904541015625,
-0.013946533203125,
-0.045684814453125,
-0.0087890625,
-0.024444580078125,
0.007251739501953125,
-0.04052734375,
-0.044158935546875,
-0.00846099853515625,
-0.014190673828125,
0.0166473388671875,
0.023040771484375,
-0.001964569091796875,
0.03857421875,
-0.072265625,
0.01197052001953125,
-0.012786865234375,
0.032867431640625,
-0.032196044921875,
-0.0535888671875,
-0.03167724609375,
0.005950927734375,
-0.04547119140625,
-0.04742431640625,
0.0367431640625,
0.01001739501953125,
0.023468017578125,
0.02703857421875,
0.0167999267578125,
0.026336669921875,
-0.05572509765625,
0.07794189453125,
-0.005634307861328125,
-0.050201416015625,
0.033905029296875,
-0.03253173828125,
0.0340576171875,
0.0723876953125,
0.0192413330078125,
-0.0263824462890625,
-0.040985107421875,
-0.0521240234375,
-0.06787109375,
0.06817626953125,
0.053680419921875,
-0.0115203857421875,
0.0193023681640625,
-0.0029544830322265625,
0.0018243789672851562,
0.01415252685546875,
-0.0850830078125,
-0.0242156982421875,
0.007457733154296875,
-0.0268402099609375,
-0.0148468017578125,
-0.022796630859375,
-0.009033203125,
-0.017242431640625,
0.083251953125,
0.0108795166015625,
0.01544189453125,
0.0295562744140625,
-0.01241302490234375,
-0.01013946533203125,
0.0246124267578125,
0.0728759765625,
0.03631591796875,
-0.0374755859375,
-0.0213775634765625,
0.0283355712890625,
-0.034820556640625,
-0.01204681396484375,
0.003551483154296875,
-0.0293426513671875,
0.0242462158203125,
0.0289764404296875,
0.0765380859375,
0.023101806640625,
-0.0496826171875,
0.03131103515625,
-0.022247314453125,
-0.03753662109375,
-0.049285888671875,
-0.01025390625,
0.00905609130859375,
0.004787445068359375,
0.0214996337890625,
0.0038433074951171875,
0.0118255615234375,
-0.01129150390625,
0.0085906982421875,
0.0032825469970703125,
-0.048614501953125,
-0.041778564453125,
0.02978515625,
0.006649017333984375,
-0.026123046875,
0.035491943359375,
-0.0283355712890625,
-0.055877685546875,
0.0297393798828125,
0.01165008544921875,
0.07470703125,
-0.0198822021484375,
-0.014007568359375,
0.05682373046875,
0.043731689453125,
-0.022216796875,
0.0369873046875,
0.005046844482421875,
-0.042999267578125,
-0.040618896484375,
-0.0638427734375,
-0.01241302490234375,
0.0104217529296875,
-0.066650390625,
0.0265960693359375,
0.029632568359375,
0.00750732421875,
-0.0294342041015625,
0.0186767578125,
-0.037261962890625,
0.00762176513671875,
-0.024139404296875,
0.08062744140625,
-0.07562255859375,
0.062744140625,
0.035980224609375,
-0.0107421875,
-0.062164306640625,
-0.01800537109375,
-0.016204833984375,
-0.0265350341796875,
0.049346923828125,
0.0210723876953125,
0.0271453857421875,
-0.01390838623046875,
-0.0200958251953125,
-0.0650634765625,
0.08624267578125,
0.0127410888671875,
-0.042694091796875,
0.00229644775390625,
0.0168609619140625,
0.0379638671875,
-0.0277252197265625,
0.002140045166015625,
0.030487060546875,
0.056396484375,
0.01120758056640625,
-0.0811767578125,
-0.030181884765625,
-0.04022216796875,
-0.0246734619140625,
0.04058837890625,
-0.03662109375,
0.06817626953125,
0.040435791015625,
-0.01003265380859375,
0.0148162841796875,
0.04266357421875,
0.0243988037109375,
0.029022216796875,
0.04058837890625,
0.0850830078125,
0.019622802734375,
-0.038116455078125,
0.0687255859375,
-0.026458740234375,
0.0360107421875,
0.08929443359375,
-0.0088348388671875,
0.06890869140625,
0.0211639404296875,
-0.00731658935546875,
0.0305023193359375,
0.044158935546875,
-0.022125244140625,
0.036865234375,
0.005947113037109375,
0.013214111328125,
-0.01032257080078125,
0.0229949951171875,
-0.0523681640625,
0.0214691162109375,
0.0188140869140625,
-0.01474761962890625,
0.00397491455078125,
0.00274658203125,
0.0002701282501220703,
-0.00910186767578125,
-0.01137542724609375,
0.05023193359375,
-0.004749298095703125,
-0.042816162109375,
0.041107177734375,
-0.01149749755859375,
0.05291748046875,
-0.056976318359375,
0.00689697265625,
0.0004763603210449219,
0.0191650390625,
0.006195068359375,
-0.03900146484375,
0.032470703125,
0.00027489662170410156,
-0.0221405029296875,
-0.0289764404296875,
0.0196380615234375,
-0.045074462890625,
-0.0740966796875,
0.029632568359375,
0.0232086181640625,
0.030792236328125,
0.005615234375,
-0.06585693359375,
0.0035648345947265625,
0.01006317138671875,
-0.046875,
0.0039520263671875,
0.05059814453125,
0.022796630859375,
0.035736083984375,
0.050323486328125,
0.019927978515625,
0.0218658447265625,
-0.0113983154296875,
0.053985595703125,
-0.0304107666015625,
-0.0325927734375,
-0.05194091796875,
0.06268310546875,
-0.012359619140625,
-0.050201416015625,
0.0565185546875,
0.07794189453125,
0.07861328125,
-0.01256561279296875,
0.023223876953125,
-0.0075836181640625,
0.05609130859375,
-0.04425048828125,
0.05108642578125,
-0.06915283203125,
0.01134490966796875,
-0.016632080078125,
-0.071044921875,
-0.01337432861328125,
0.029296875,
-0.01422882080078125,
-0.0303802490234375,
0.051971435546875,
0.047149658203125,
-0.0136566162109375,
-0.0152130126953125,
0.0241241455078125,
0.028778076171875,
0.0145416259765625,
0.037445068359375,
0.025634765625,
-0.06915283203125,
0.04278564453125,
-0.02484130859375,
-0.002643585205078125,
-0.005001068115234375,
-0.05438232421875,
-0.061187744140625,
-0.047698974609375,
-0.0086669921875,
-0.0144805908203125,
-0.01947021484375,
0.06439208984375,
0.03887939453125,
-0.06939697265625,
-0.03668212890625,
0.005603790283203125,
0.004840850830078125,
-0.01348114013671875,
-0.0189666748046875,
0.0469970703125,
-0.02435302734375,
-0.068603515625,
0.03521728515625,
0.00949859619140625,
-0.006435394287109375,
-0.003765106201171875,
-0.025299072265625,
-0.03759765625,
-0.0009756088256835938,
0.0267181396484375,
-0.0012521743774414062,
-0.043792724609375,
0.006114959716796875,
0.0164947509765625,
-0.003143310546875,
0.0261383056640625,
0.030792236328125,
-0.016937255859375,
0.0246429443359375,
0.062255859375,
0.0233917236328125,
0.027435302734375,
-0.00945281982421875,
0.04302978515625,
-0.046600341796875,
0.0216217041015625,
0.018463134765625,
0.03900146484375,
0.0230865478515625,
-0.006595611572265625,
0.06097412109375,
0.0167999267578125,
-0.05157470703125,
-0.08465576171875,
0.0027828216552734375,
-0.0894775390625,
0.002582550048828125,
0.07000732421875,
-0.0255889892578125,
-0.0198822021484375,
0.0192108154296875,
-0.013427734375,
0.0068206787109375,
-0.0207366943359375,
0.02923583984375,
0.076171875,
0.0274200439453125,
0.015899658203125,
-0.059967041015625,
0.024200439453125,
0.03680419921875,
-0.048126220703125,
-0.01245880126953125,
0.011810302734375,
0.011260986328125,
0.0290069580078125,
0.0309906005859375,
-0.0236663818359375,
0.007495880126953125,
-0.0230865478515625,
0.02825927734375,
0.0003535747528076172,
-0.01800537109375,
-0.0210418701171875,
-0.006092071533203125,
-0.01334381103515625,
-0.016448974609375
]
] |
facebook/blenderbot-3B | 2023-03-30T16:12:22.000Z | [
"transformers",
"pytorch",
"blenderbot",
"text2text-generation",
"convAI",
"conversational",
"facebook",
"en",
"dataset:blended_skill_talk",
"arxiv:1907.06616",
"license:apache-2.0",
"autotrain_compatible",
"endpoints_compatible",
"has_space",
"region:us"
] | conversational | facebook | null | null | facebook/blenderbot-3B | 113 | 7,289 | transformers | 2022-03-02T23:29:05 | ---
language:
- en
thumbnail:
tags:
- convAI
- conversational
- facebook
license: apache-2.0
datasets:
- blended_skill_talk
metrics:
- perplexity
---
## Model description
+ Paper: [Recipes for building an open-domain chatbot](https://arxiv.org/abs/1907.06616)
+ [Original PARLAI Code](https://parl.ai/projects/recipes/)
### Abstract
Building open-domain chatbots is a challenging area for machine learning research. While prior work has shown that scaling neural models in the number of parameters and the size of the data they are trained on gives improved results, we show that other ingredients are important for a high-performing chatbot. Good conversation requires a number of skills that an expert conversationalist blends in a seamless way: providing engaging talking points and listening to their partners, both asking and answering questions, and displaying knowledge, empathy and personality appropriately, depending on the situation. We show that large scale models can learn these skills when given appropriate training data and choice of generation strategy. We build variants of these recipes with 90M, 2.7B and 9.4B parameter neural models, and make our models and code publicly available. Human evaluations show our best models are superior to existing approaches in multi-turn dialogue in terms of engagingness and humanness measurements. We then discuss the limitations of this work by analyzing failure cases of our models.
| 1,451 | [
[
-0.0259857177734375,
-0.06011962890625,
0.025848388671875,
0.0245361328125,
0.0245208740234375,
-0.006683349609375,
-0.03216552734375,
-0.0158843994140625,
-0.0006651878356933594,
0.050262451171875,
-0.02069091796875,
-0.0195465087890625,
-0.057037353515625,
-0.0283966064453125,
-0.017578125,
0.072998046875,
0.0250396728515625,
0.01373291015625,
-0.025238037109375,
-0.0140228271484375,
-0.036529541015625,
-0.05584716796875,
-0.061492919921875,
-0.01462554931640625,
0.047576904296875,
0.035797119140625,
0.07159423828125,
0.035430908203125,
0.014190673828125,
0.02496337890625,
-0.024658203125,
0.035736083984375,
-0.056549072265625,
0.02325439453125,
-0.002819061279296875,
-0.030517578125,
-0.06591796875,
0.017974853515625,
0.004482269287109375,
0.0634765625,
-0.0253143310546875,
0.00525665283203125,
0.0166778564453125,
0.04119873046875,
-0.0301513671875,
0.037994384765625,
-0.0634765625,
-0.016632080078125,
0.0071868896484375,
-0.01342010498046875,
-0.058685302734375,
-0.01406097412109375,
0.0187835693359375,
-0.04595947265625,
0.016204833984375,
-0.00466156005859375,
0.052947998046875,
-0.0024127960205078125,
-0.0213470458984375,
-0.0279998779296875,
-0.07220458984375,
0.055206298828125,
-0.068115234375,
0.01517486572265625,
0.045318603515625,
0.036224365234375,
-0.033111572265625,
-0.055816650390625,
-0.031524658203125,
-0.0220947265625,
0.00896453857421875,
-0.0059051513671875,
-0.01641845703125,
-0.0004105567932128906,
0.006855010986328125,
0.016571044921875,
-0.0211334228515625,
0.01337432861328125,
-0.0245208740234375,
0.00925445556640625,
0.045562744140625,
0.031829833984375,
0.0182037353515625,
0.01210784912109375,
-0.002384185791015625,
-0.01500701904296875,
-0.03204345703125,
0.006618499755859375,
0.058258056640625,
0.04608154296875,
-0.02044677734375,
0.041717529296875,
-0.01270294189453125,
0.050628662109375,
-0.0012865066528320312,
-0.00621795654296875,
-0.004741668701171875,
-0.0307159423828125,
-0.00624847412109375,
-0.03253173828125,
0.06365966796875,
0.0322265625,
0.037384033203125,
-0.00824737548828125,
0.0168914794921875,
-0.028961181640625,
0.04681396484375,
-0.0771484375,
-0.035003662109375,
0.0286865234375,
-0.045806884765625,
-0.025848388671875,
-0.00528717041015625,
-0.042816162109375,
-0.0246124267578125,
-0.019866943359375,
0.0172576904296875,
-0.046356201171875,
-0.036651611328125,
0.0208282470703125,
-0.0164794921875,
0.01024627685546875,
0.0322265625,
-0.0887451171875,
0.023345947265625,
0.056610107421875,
0.06378173828125,
0.01436614990234375,
-0.0223846435546875,
-0.034210205078125,
-0.0247955322265625,
-0.036041259765625,
0.0303802490234375,
-0.045806884765625,
-0.0242919921875,
0.00748443603515625,
-0.01349639892578125,
-0.0007543563842773438,
-0.038543701171875,
0.034576416015625,
-0.031982421875,
0.029754638671875,
-0.0050048828125,
-0.045318603515625,
-0.0113067626953125,
0.007740020751953125,
-0.039459228515625,
0.0352783203125,
0.014678955078125,
-0.043609619140625,
0.0019588470458984375,
-0.07037353515625,
-0.036102294921875,
0.01297760009765625,
0.0014171600341796875,
-0.0217437744140625,
0.0228271484375,
0.0164794921875,
0.04815673828125,
-0.026214599609375,
0.0238494873046875,
-0.028839111328125,
0.01201629638671875,
0.04150390625,
-0.037353515625,
0.0677490234375,
0.009185791015625,
0.005756378173828125,
0.0285186767578125,
-0.045501708984375,
0.02850341796875,
0.0203094482421875,
-0.021331787109375,
-0.01088714599609375,
0.0091705322265625,
0.004421234130859375,
-0.00774383544921875,
-0.0022869110107421875,
-0.0300140380859375,
0.00014472007751464844,
-0.03875732421875,
0.062255859375,
0.047332763671875,
-0.004566192626953125,
0.0380859375,
-0.0268707275390625,
0.0191802978515625,
0.0089263916015625,
0.02239990234375,
-0.037261962890625,
-0.07232666015625,
-0.06890869140625,
-0.029388427734375,
0.037200927734375,
0.0259857177734375,
-0.033447265625,
0.06500244140625,
0.005649566650390625,
-0.072021484375,
-0.0679931640625,
-0.004993438720703125,
0.0301971435546875,
0.034332275390625,
0.0178985595703125,
-0.03497314453125,
-0.034393310546875,
-0.060577392578125,
-0.0170745849609375,
-0.0013990402221679688,
-0.021728515625,
0.0528564453125,
0.032073974609375,
0.0119781494140625,
0.06689453125,
-0.04058837890625,
0.0025386810302734375,
-0.0316162109375,
0.01438140869140625,
0.01267242431640625,
0.034149169921875,
0.028656005859375,
-0.06622314453125,
-0.05377197265625,
0.00817108154296875,
-0.062469482421875,
0.0277099609375,
-0.0011644363403320312,
-0.026763916015625,
-0.0011835098266601562,
0.0140380859375,
-0.07281494140625,
0.040618896484375,
0.027679443359375,
-0.018402099609375,
0.027130126953125,
-0.01873779296875,
0.01495361328125,
-0.09014892578125,
0.000934600830078125,
0.01023101806640625,
0.00490570068359375,
-0.07861328125,
0.0092620849609375,
0.006725311279296875,
-0.0307464599609375,
-0.048309326171875,
0.04412841796875,
-0.01424407958984375,
0.0146026611328125,
-0.0158233642578125,
-0.00940704345703125,
-0.00537109375,
0.0635986328125,
0.0020904541015625,
0.054931640625,
0.0362548828125,
-0.037628173828125,
0.0251007080078125,
0.0367431640625,
-0.0215606689453125,
0.03887939453125,
-0.08697509765625,
0.01763916015625,
0.01464080810546875,
0.0113525390625,
-0.07427978515625,
-0.042816162109375,
0.0004353523254394531,
-0.0628662109375,
0.01241302490234375,
-0.024444580078125,
-0.041046142578125,
-0.00972747802734375,
0.01078033447265625,
0.002231597900390625,
0.056243896484375,
-0.033782958984375,
0.04486083984375,
0.0377197265625,
-0.0184783935546875,
-0.007080078125,
-0.007350921630859375,
0.007495880126953125,
-0.004730224609375,
-0.062408447265625,
0.01360321044921875,
-0.0253448486328125,
-0.00782012939453125,
-0.0139312744140625,
0.023101806640625,
-0.0169830322265625,
0.007312774658203125,
0.028900146484375,
0.0126953125,
-0.0212860107421875,
-0.01019287109375,
-0.010498046875,
-0.005558013916015625,
0.00926971435546875,
-0.022705078125,
0.058197021484375,
-0.021331787109375,
-0.0094451904296875,
-0.0433349609375,
0.036346435546875,
0.04888916015625,
-0.025634765625,
0.07427978515625,
0.033935546875,
-0.017303466796875,
-0.01177978515625,
-0.02313232421875,
-0.0277252197265625,
-0.034423828125,
0.0210418701171875,
-0.015289306640625,
-0.06024169921875,
0.037628173828125,
0.0134124755859375,
0.01116180419921875,
0.0177001953125,
0.0660400390625,
0.0010881423950195312,
0.09002685546875,
0.0419921875,
0.0173187255859375,
0.0210723876953125,
-0.007259368896484375,
0.0263214111328125,
-0.0465087890625,
-0.0215911865234375,
-0.046661376953125,
-0.00936126708984375,
-0.0308685302734375,
-0.037841796875,
0.02587890625,
-0.011932373046875,
-0.034454345703125,
0.037994384765625,
-0.0311279296875,
0.04119873046875,
0.07086181640625,
0.0261993408203125,
-0.0014467239379882812,
-0.0066375732421875,
0.0028057098388671875,
-0.0018224716186523438,
-0.06402587890625,
-0.0433349609375,
0.0869140625,
0.0362548828125,
0.048583984375,
-0.006000518798828125,
0.0277862548828125,
-0.00787353515625,
0.0004425048828125,
-0.06964111328125,
0.03973388671875,
0.006561279296875,
-0.0657958984375,
-0.03668212890625,
-0.04168701171875,
-0.06939697265625,
0.00726318359375,
-0.0123443603515625,
-0.034759521484375,
-0.0243988037109375,
0.00959014892578125,
-0.025848388671875,
0.0288848876953125,
-0.07257080078125,
0.08038330078125,
-0.00940704345703125,
-0.0221405029296875,
-0.001094818115234375,
-0.06640625,
0.01288604736328125,
0.021575927734375,
-0.01336669921875,
-0.00409698486328125,
0.020904541015625,
0.0330810546875,
-0.0277557373046875,
0.08984375,
-0.01345062255859375,
0.0039520263671875,
0.0287322998046875,
0.0175018310546875,
0.00984954833984375,
0.0021648406982421875,
0.0014600753784179688,
0.0230255126953125,
-0.0235748291015625,
-0.04754638671875,
-0.058258056640625,
0.04022216796875,
-0.07415771484375,
-0.040771484375,
0.00531768798828125,
-0.05029296875,
-0.026763916015625,
0.0098114013671875,
0.00909423828125,
0.033966064453125,
-0.03466796875,
0.05462646484375,
0.066650390625,
-0.03594970703125,
0.017242431640625,
0.022003173828125,
0.010711669921875,
-0.03924560546875,
0.0595703125,
-0.004573822021484375,
0.037139892578125,
0.023590087890625,
0.0238037109375,
0.0195159912109375,
-0.02386474609375,
-0.0300140380859375,
-0.00957489013671875,
-0.033935546875,
-0.011993408203125,
-0.05706787109375,
-0.04522705078125,
-0.03668212890625,
-0.006702423095703125,
-0.055267333984375,
-0.02935791015625,
-0.031494140625,
0.00821685791015625,
0.0310211181640625,
0.047454833984375,
0.00919342041015625,
0.0291748046875,
-0.06329345703125,
0.004878997802734375,
0.0235595703125,
0.025634765625,
0.042816162109375,
-0.036041259765625,
-0.0260772705078125,
0.0238800048828125,
-0.046875,
-0.032318115234375,
0.049041748046875,
0.01010894775390625,
0.039459228515625,
0.0193023681640625,
-0.0028705596923828125,
0.024322509765625,
-0.04449462890625,
0.06585693359375,
0.028656005859375,
-0.05426025390625,
0.046295166015625,
-0.044647216796875,
0.0206756591796875,
0.022613525390625,
0.06744384765625,
-0.03350830078125,
-0.024444580078125,
-0.058380126953125,
-0.05804443359375,
0.050048828125,
0.03955078125,
0.048797607421875,
-0.0084686279296875,
0.0191497802734375,
0.03466796875,
0.0182342529296875,
-0.0282440185546875,
-0.01123046875,
-0.03216552734375,
-0.01546478271484375,
-0.0055084228515625,
-0.006397247314453125,
-0.0217132568359375,
-0.022369384765625,
0.032470703125,
-0.00627899169921875,
0.0347900390625,
-0.024322509765625,
0.03363037109375,
0.001338958740234375,
0.01055908203125,
0.044281005859375,
0.0455322265625,
-0.0146026611328125,
-0.01397705078125,
-0.0187530517578125,
-0.0082244873046875,
-0.016845703125,
-0.0148468017578125,
0.02984619140625,
-0.0321044921875,
0.0301513671875,
0.0718994140625,
0.0035343170166015625,
-0.04962158203125,
0.042449951171875,
-0.02386474609375,
-0.03314208984375,
-0.005359649658203125,
0.03216552734375,
0.0419921875,
0.026397705078125,
0.006801605224609375,
0.0190582275390625,
-0.01503753662109375,
-0.05126953125,
0.003154754638671875,
0.0197296142578125,
-0.041351318359375,
-0.040618896484375,
0.044647216796875,
0.0338134765625,
-0.06365966796875,
0.0704345703125,
-0.004291534423828125,
-0.0285186767578125,
0.037567138671875,
0.040008544921875,
0.06646728515625,
-0.014678955078125,
0.0270233154296875,
0.01910400390625,
0.0020751953125,
-0.0032596588134765625,
0.0127105712890625,
-0.007106781005859375,
-0.0577392578125,
-0.00980377197265625,
-0.026947021484375,
-0.06390380859375,
0.007518768310546875,
-0.0394287109375,
0.007373809814453125,
-0.032470703125,
-0.01251220703125,
0.03009033203125,
-0.030792236328125,
-0.05804443359375,
-0.00888824462890625,
-0.01491546630859375,
0.061187744140625,
-0.038238525390625,
0.04644775390625,
0.02764892578125,
-0.043975830078125,
-0.045928955078125,
-0.0220947265625,
-0.0167236328125,
-0.058380126953125,
0.03778076171875,
-0.0008296966552734375,
0.0242156982421875,
-0.0199432373046875,
-0.076171875,
-0.031707763671875,
0.055999755859375,
0.0257110595703125,
-0.018890380859375,
-0.033355712890625,
0.00374603271484375,
0.060150146484375,
-0.05804443359375,
0.0284576416015625,
0.01090240478515625,
0.0034885406494140625,
0.0302734375,
-0.08294677734375,
-0.01544189453125,
-0.0243682861328125,
0.007221221923828125,
0.00560760498046875,
-0.049530029296875,
0.071533203125,
-0.027130126953125,
0.018646240234375,
0.0229034423828125,
0.055206298828125,
0.000014841556549072266,
0.02178955078125,
0.0164642333984375,
0.032318115234375,
0.031494140625,
0.0036869049072265625,
0.05743408203125,
-0.0213623046875,
0.0055999755859375,
0.10516357421875,
-0.0269012451171875,
0.08056640625,
0.01116180419921875,
0.0027675628662109375,
0.027252197265625,
0.03521728515625,
0.00925445556640625,
0.0318603515625,
0.009521484375,
-0.004222869873046875,
-0.023284912109375,
0.0014171600341796875,
-0.0106353759765625,
0.0498046875,
0.0364990234375,
-0.03778076171875,
0.0000017285346984863281,
0.0017232894897460938,
0.01395416259765625,
0.007251739501953125,
0.0145111083984375,
0.07177734375,
0.007335662841796875,
-0.06317138671875,
0.047760009765625,
-0.0073089599609375,
0.017852783203125,
-0.033416748046875,
-0.00787353515625,
-0.0212860107421875,
0.01519012451171875,
0.007843017578125,
-0.058990478515625,
0.01192474365234375,
-0.0091552734375,
-0.00743865966796875,
-0.0202178955078125,
0.042572021484375,
-0.02655029296875,
-0.0087432861328125,
0.01352691650390625,
0.052642822265625,
0.004985809326171875,
-0.0184783935546875,
-0.04937744140625,
-0.006374359130859375,
-0.01476287841796875,
-0.0177764892578125,
0.0306854248046875,
0.052337646484375,
0.0235748291015625,
0.047607421875,
0.0462646484375,
-0.0034637451171875,
-0.01096343994140625,
-0.00418853759765625,
0.07305908203125,
-0.044586181640625,
-0.0262908935546875,
-0.044921875,
0.0576171875,
-0.028289794921875,
-0.044891357421875,
0.056304931640625,
0.035308837890625,
0.067138671875,
-0.0144805908203125,
0.05352783203125,
-0.005283355712890625,
0.043426513671875,
-0.021697998046875,
0.0556640625,
-0.0229034423828125,
-0.00927734375,
0.005741119384765625,
-0.061798095703125,
-0.0182342529296875,
0.036895751953125,
0.0022754669189453125,
0.01097869873046875,
0.032318115234375,
0.068115234375,
-0.013275146484375,
0.028900146484375,
0.05255126953125,
0.01082611083984375,
0.04083251953125,
0.04229736328125,
0.0670166015625,
-0.0291748046875,
0.041412353515625,
0.00551605224609375,
-0.03802490234375,
-0.0254669189453125,
-0.052642822265625,
-0.10467529296875,
-0.068359375,
-0.01641845703125,
-0.039794921875,
0.00720977783203125,
0.076416015625,
0.09429931640625,
-0.044036865234375,
-0.02972412109375,
-0.0039215087890625,
-0.0035991668701171875,
-0.019012451171875,
-0.0129241943359375,
-0.01285552978515625,
-0.04083251953125,
-0.056732177734375,
0.0343017578125,
0.0145111083984375,
0.0047149658203125,
-0.0266876220703125,
-0.01385498046875,
0.0002951622009277344,
0.040679931640625,
0.062164306640625,
0.023040771484375,
-0.0440673828125,
-0.005626678466796875,
0.00789642333984375,
-0.00681304931640625,
0.01499176025390625,
0.062164306640625,
-0.0178985595703125,
0.039093017578125,
0.034088134765625,
0.056915283203125,
0.038177490234375,
-0.0205841064453125,
0.062744140625,
-0.05322265625,
-0.002597808837890625,
0.0177001953125,
0.01312255859375,
0.0236663818359375,
-0.00841522216796875,
0.039306640625,
-0.007312774658203125,
-0.06451416015625,
-0.0535888671875,
0.0206756591796875,
-0.07373046875,
-0.01763916015625,
0.0804443359375,
-0.0225067138671875,
-0.017608642578125,
0.0111083984375,
-0.0287322998046875,
0.0179443359375,
-0.04156494140625,
0.0552978515625,
0.057891845703125,
-0.01195526123046875,
0.008514404296875,
-0.059661865234375,
0.0347900390625,
0.0035800933837890625,
-0.060760498046875,
0.033172607421875,
0.024871826171875,
-0.0014715194702148438,
0.01529693603515625,
0.033935546875,
-0.01273345947265625,
0.0012407302856445312,
0.0254669189453125,
0.006793975830078125,
-0.0100555419921875,
-0.052001953125,
0.00823974609375,
0.033599853515625,
0.0179595947265625,
-0.0029201507568359375
]
] |
wavymulder/portraitplus | 2023-05-05T21:59:07.000Z | [
"diffusers",
"stable-diffusion",
"stable-diffusion-diffusers",
"text-to-image",
"safetensors",
"en",
"license:creativeml-openrail-m",
"endpoints_compatible",
"has_space",
"diffusers:StableDiffusionPipeline",
"region:us"
] | text-to-image | wavymulder | null | null | wavymulder/portraitplus | 289 | 7,284 | diffusers | 2022-12-23T16:04:26 | ---
language:
- en
thumbnail: "https://huggingface.co/wavymulder/portraitplus/resolve/main/imgs/page1.jpg"
license: creativeml-openrail-m
tags:
- stable-diffusion
- stable-diffusion-diffusers
- text-to-image
- safetensors
- diffusers
inference: true
---
**Portrait+**
[*CKPT DOWNLOAD LINK*](https://huggingface.co/wavymulder/portraitplus/resolve/main/portrait%2B1.0.ckpt) - this is a dreambooth model trained on a diverse set of close to medium range portraits of people.
Use `portrait+ style` in your prompt (I recommend at the start)
The goal was to create a model with a consistent portrait composition and consistent eyes. See the batch example below for the consistency of the model's eyes. This model can do several styles, so you'll want to guide it along depending on your goals. Note below in the document that prompting celebrities works a bit differently than prompting generic characters, since real people have a more photoreal presence in the base 1.5 model. Also note that fantasy concepts, like cyberpunk people or wizards, will require more rigid prompting for photoreal styles than something common like a person in a park.
Portrait+ works best at a 1:1 aspect ratio, though I've had success with tall aspect ratios as well.
Please see [this document where I share the parameters (prompt, sampler, seed, etc.) used for all example images above.](https://huggingface.co/wavymulder/portraitplus/resolve/main/parameters_for_samples.txt)
We support a [Gradio](https://github.com/gradio-app/gradio) Web UI to run portraitplus:
[](https://huggingface.co/spaces/wavymulder/portraitplus)
 | 2,151 | [
[
-0.051177978515625,
-0.035736083984375,
0.027435302734375,
0.025177001953125,
-0.028564453125,
0.0002903938293457031,
0.0013685226440429688,
-0.051177978515625,
0.05291748046875,
0.044830322265625,
-0.06884765625,
-0.036895751953125,
-0.0266571044921875,
-0.0010738372802734375,
0.01383209228515625,
0.07171630859375,
-0.0184783935546875,
-0.02569580078125,
-0.0192413330078125,
0.0001513957977294922,
-0.052001953125,
0.01702880859375,
-0.057952880859375,
-0.01290130615234375,
0.01763916015625,
0.040374755859375,
0.0732421875,
0.0305938720703125,
0.029327392578125,
0.0189208984375,
0.0051116943359375,
0.019500732421875,
-0.036102294921875,
0.01806640625,
-0.0115203857421875,
-0.0230560302734375,
-0.0377197265625,
0.03143310546875,
0.054901123046875,
0.016998291015625,
-0.01641845703125,
0.038299560546875,
-0.01306915283203125,
0.06695556640625,
-0.0234527587890625,
0.0158538818359375,
-0.00299835205078125,
0.00452423095703125,
-0.0223388671875,
-0.003147125244140625,
-0.005825042724609375,
-0.037078857421875,
-0.017303466796875,
-0.065185546875,
0.038848876953125,
0.0284576416015625,
0.079345703125,
-0.002071380615234375,
-0.01284027099609375,
0.03271484375,
-0.04437255859375,
0.02899169921875,
-0.0186920166015625,
0.0272216796875,
0.01438140869140625,
0.05987548828125,
-0.01953125,
-0.035736083984375,
-0.04559326171875,
-0.024688720703125,
-0.0003261566162109375,
0.010711669921875,
-0.0076141357421875,
0.0110931396484375,
0.039398193359375,
0.0347900390625,
-0.05950927734375,
-0.0193634033203125,
-0.0294036865234375,
-0.003955841064453125,
0.04156494140625,
0.02252197265625,
0.04107666015625,
-0.00228118896484375,
-0.028045654296875,
-0.03839111328125,
-0.037109375,
0.042083740234375,
0.0360107421875,
0.00823974609375,
-0.045379638671875,
0.0411376953125,
-0.0152587890625,
0.059326171875,
0.0433349609375,
-0.019744873046875,
0.024749755859375,
-0.00055694580078125,
0.001476287841796875,
-0.0325927734375,
0.044708251953125,
0.07269287109375,
0.0075225830078125,
0.007587432861328125,
-0.01473236083984375,
0.00484466552734375,
0.020599365234375,
-0.0736083984375,
-0.007762908935546875,
0.0386962890625,
-0.035369873046875,
-0.01953125,
-0.0052032470703125,
-0.068359375,
-0.030853271484375,
-0.01947021484375,
0.017791748046875,
-0.034423828125,
-0.022430419921875,
-0.0012063980102539062,
-0.0335693359375,
0.017333984375,
0.04229736328125,
-0.07489013671875,
0.0163116455078125,
0.0198974609375,
0.0535888671875,
-0.0024242401123046875,
-0.0154266357421875,
-0.032623291015625,
0.01250457763671875,
-0.023651123046875,
0.053741455078125,
-0.0206146240234375,
-0.0254669189453125,
-0.002025604248046875,
0.00661468505859375,
-0.01258087158203125,
-0.03997802734375,
0.0218658447265625,
-0.0229644775390625,
0.0261688232421875,
-0.0222930908203125,
-0.0201568603515625,
-0.0207061767578125,
0.0120391845703125,
-0.052215576171875,
0.047210693359375,
0.032867431640625,
-0.05072021484375,
0.011688232421875,
-0.062042236328125,
-0.0179901123046875,
0.0413818359375,
-0.040863037109375,
-0.018524169921875,
0.01116180419921875,
0.0054779052734375,
0.036346435546875,
-0.006496429443359375,
-0.01065826416015625,
-0.024322509765625,
-0.0300140380859375,
-0.00945281982421875,
0.02178955078125,
0.07537841796875,
0.01351165771484375,
-0.0247039794921875,
0.0135955810546875,
-0.043792724609375,
-0.00101470947265625,
0.04522705078125,
-0.0022754669189453125,
0.01396942138671875,
-0.034210205078125,
0.038299560546875,
0.0135650634765625,
0.027740478515625,
-0.0517578125,
0.032806396484375,
0.0183868408203125,
-0.004474639892578125,
0.040802001953125,
0.005352020263671875,
0.01078033447265625,
-0.0477294921875,
0.0771484375,
0.0020999908447265625,
0.0199432373046875,
-0.016326904296875,
-0.040557861328125,
-0.029876708984375,
-0.025665283203125,
-0.0240631103515625,
0.037628173828125,
-0.054351806640625,
0.00513458251953125,
-0.00627899169921875,
-0.0499267578125,
-0.0295257568359375,
-0.002643585205078125,
0.03839111328125,
0.035888671875,
-0.00818634033203125,
-0.056884765625,
-0.018402099609375,
-0.0760498046875,
-0.024566650390625,
-0.0284423828125,
0.0017633438110351562,
0.038055419921875,
0.017974853515625,
-0.016571044921875,
0.0592041015625,
-0.034637451171875,
-0.0161895751953125,
-0.005558013916015625,
-0.01381683349609375,
0.035858154296875,
0.05963134765625,
0.0748291015625,
-0.06134033203125,
-0.0255889892578125,
-0.003025054931640625,
-0.048492431640625,
0.01277923583984375,
-0.006175994873046875,
-0.01351165771484375,
-0.005767822265625,
-0.0025482177734375,
-0.07843017578125,
0.02093505859375,
0.0282745361328125,
-0.052520751953125,
0.07586669921875,
-0.002841949462890625,
0.03582763671875,
-0.07696533203125,
0.0223541259765625,
0.04168701171875,
-0.04730224609375,
-0.0400390625,
0.0501708984375,
-0.0172576904296875,
-0.0288238525390625,
-0.034759521484375,
0.054656982421875,
-0.0306854248046875,
0.0172271728515625,
-0.0029296875,
-0.0167083740234375,
0.003101348876953125,
0.027679443359375,
0.0016584396362304688,
0.044830322265625,
0.052276611328125,
-0.023040771484375,
0.052337646484375,
0.033721923828125,
-0.035888671875,
0.06365966796875,
-0.053741455078125,
0.02606201171875,
-0.03094482421875,
0.0242767333984375,
-0.09051513671875,
-0.06695556640625,
0.046112060546875,
-0.029876708984375,
0.032958984375,
-0.0022754669189453125,
-0.038787841796875,
-0.050048828125,
-0.026702880859375,
0.033355712890625,
0.06573486328125,
-0.03326416015625,
0.05364990234375,
0.0185394287109375,
-0.01174163818359375,
-0.013885498046875,
-0.0230712890625,
0.007152557373046875,
-0.0294647216796875,
-0.065673828125,
0.046356201171875,
-0.006237030029296875,
-0.0223846435546875,
-0.0142669677734375,
0.00557708740234375,
0.0037212371826171875,
-0.033355712890625,
0.04058837890625,
0.04986572265625,
-0.00809478759765625,
-0.023590087890625,
-0.0035572052001953125,
-0.002719879150390625,
-0.00743865966796875,
0.005924224853515625,
0.0254974365234375,
-0.057647705078125,
-0.01332855224609375,
-0.08416748046875,
0.021453857421875,
0.0634765625,
0.00620269775390625,
0.03631591796875,
0.054473876953125,
-0.028717041015625,
0.0179443359375,
-0.06396484375,
-0.0201263427734375,
-0.03656005859375,
-0.0004639625549316406,
-0.036712646484375,
-0.052642822265625,
0.056549072265625,
-0.0008502006530761719,
0.036956787109375,
0.031494140625,
0.044464111328125,
-0.020111083984375,
0.0650634765625,
0.044403076171875,
0.015716552734375,
0.053863525390625,
-0.0205535888671875,
-0.019134521484375,
-0.0474853515625,
-0.030670166015625,
-0.0004394054412841797,
-0.0396728515625,
-0.034912109375,
-0.0168609619140625,
0.024383544921875,
0.00963592529296875,
-0.0235443115234375,
0.060882568359375,
-0.0367431640625,
0.04180908203125,
0.03839111328125,
0.0289459228515625,
0.0101470947265625,
0.0193023681640625,
-0.019989013671875,
0.0010318756103515625,
-0.01971435546875,
-0.035430908203125,
0.0249481201171875,
0.0272216796875,
0.0638427734375,
0.007427215576171875,
0.06549072265625,
-0.0027866363525390625,
0.016845703125,
-0.040863037109375,
0.056427001953125,
0.003993988037109375,
-0.07177734375,
-0.01153564453125,
-0.00537109375,
-0.04608154296875,
0.0257415771484375,
-0.0184173583984375,
-0.07830810546875,
0.03594970703125,
0.03765869140625,
-0.049713134765625,
0.01021575927734375,
-0.05535888671875,
0.07171630859375,
-0.0018787384033203125,
-0.029876708984375,
0.0025730133056640625,
-0.050933837890625,
0.031494140625,
0.0164794921875,
-0.00571441650390625,
-0.0401611328125,
0.01136016845703125,
0.046875,
-0.03271484375,
0.06451416015625,
-0.047332763671875,
0.0135345458984375,
0.0249786376953125,
0.01947021484375,
0.040496826171875,
-0.0013179779052734375,
0.0033588409423828125,
-0.003856658935546875,
0.028900146484375,
-0.03216552734375,
-0.0396728515625,
0.02801513671875,
-0.0496826171875,
-0.058746337890625,
-0.00855255126953125,
-0.051483154296875,
-0.0130767822265625,
0.0064544677734375,
0.047698974609375,
0.0269012451171875,
-0.01424407958984375,
0.01291656494140625,
0.0193634033203125,
0.0065155029296875,
0.0195159912109375,
0.00780487060546875,
-0.0546875,
-0.03271484375,
0.0447998046875,
-0.018280029296875,
0.017730712890625,
0.00009304285049438477,
0.01885986328125,
-0.018280029296875,
-0.01230621337890625,
-0.05914306640625,
0.0241851806640625,
-0.037322998046875,
0.000988006591796875,
-0.0101318359375,
0.0011720657348632812,
-0.0428466796875,
-0.0240325927734375,
-0.032562255859375,
-0.03955078125,
-0.0231475830078125,
0.009124755859375,
0.05316162109375,
0.0134124755859375,
0.004817962646484375,
0.0286712646484375,
-0.04010009765625,
0.0478515625,
0.0294647216796875,
0.009490966796875,
-0.00951385498046875,
-0.0400390625,
0.0132904052734375,
0.0012865066528320312,
-0.034515380859375,
-0.05419921875,
0.034149169921875,
0.01093292236328125,
0.0399169921875,
0.044189453125,
-0.0037212371826171875,
0.077392578125,
-0.0277557373046875,
0.051971435546875,
0.047210693359375,
-0.02850341796875,
0.056915283203125,
-0.072265625,
0.028106689453125,
0.07501220703125,
0.0275421142578125,
-0.06219482421875,
-0.0231781005859375,
-0.05902099609375,
-0.0269927978515625,
0.043853759765625,
0.021209716796875,
0.0225830078125,
0.0165252685546875,
0.05462646484375,
0.01108551025390625,
0.01490020751953125,
-0.057952880859375,
-0.0244903564453125,
-0.01465606689453125,
0.002246856689453125,
0.002002716064453125,
0.010955810546875,
-0.0203704833984375,
-0.04681396484375,
0.04852294921875,
-0.0082550048828125,
0.0305023193359375,
0.0206451416015625,
0.038116455078125,
-0.03729248046875,
-0.01509857177734375,
0.037322998046875,
0.05975341796875,
-0.030242919921875,
-0.017822265625,
-0.0249481201171875,
-0.046600341796875,
0.013885498046875,
0.011962890625,
-0.0238800048828125,
0.00771331787109375,
0.0090179443359375,
0.07952880859375,
-0.0252532958984375,
-0.037200927734375,
0.036865234375,
-0.00168609619140625,
-0.0139923095703125,
-0.0300445556640625,
0.0218658447265625,
0.00637054443359375,
0.039459228515625,
0.020294189453125,
0.0254058837890625,
-0.0004723072052001953,
-0.039825439453125,
-0.009735107421875,
0.0285491943359375,
-0.03521728515625,
-0.0221405029296875,
0.062286376953125,
0.005680084228515625,
-0.039642333984375,
0.044891357421875,
-0.0401611328125,
-0.015625,
0.06695556640625,
0.05780029296875,
0.07110595703125,
-0.02197265625,
0.04071044921875,
0.03131103515625,
0.00833892822265625,
-0.006244659423828125,
0.0533447265625,
0.005710601806640625,
-0.043609619140625,
-0.0020885467529296875,
-0.040924072265625,
-0.03802490234375,
0.01727294921875,
-0.040252685546875,
0.0288848876953125,
-0.05230712890625,
-0.0218963623046875,
0.01363372802734375,
0.006961822509765625,
-0.06573486328125,
0.0117034912109375,
0.0028400421142578125,
0.06781005859375,
-0.0592041015625,
0.033477783203125,
0.07757568359375,
-0.0291900634765625,
-0.06787109375,
-0.005718231201171875,
0.0311279296875,
-0.06585693359375,
0.024871826171875,
0.04205322265625,
-0.0102996826171875,
0.004467010498046875,
-0.060760498046875,
-0.037384033203125,
0.0859375,
0.0396728515625,
-0.049896240234375,
0.0016012191772460938,
-0.0174713134765625,
0.0270233154296875,
-0.061676025390625,
-0.01192474365234375,
0.0170745849609375,
0.033782958984375,
0.05206298828125,
-0.068115234375,
0.0282135009765625,
-0.034881591796875,
0.0167999267578125,
-0.00801849365234375,
-0.06494140625,
0.053436279296875,
-0.037078857421875,
-0.01348876953125,
0.037841796875,
0.0384521484375,
0.04595947265625,
0.0175628662109375,
0.05364990234375,
0.048065185546875,
0.04254150390625,
0.01461029052734375,
0.0921630859375,
-0.0234375,
0.0189666748046875,
0.067626953125,
-0.004673004150390625,
0.043121337890625,
0.0224151611328125,
-0.025115966796875,
0.051666259765625,
0.0845947265625,
-0.0158538818359375,
0.032318115234375,
-0.0008425712585449219,
-0.02984619140625,
-0.0176544189453125,
-0.01558685302734375,
-0.02655029296875,
0.0280914306640625,
-0.000518798828125,
-0.0212554931640625,
-0.00738525390625,
0.0232696533203125,
-0.010833740234375,
-0.0211944580078125,
-0.0217742919921875,
0.0210418701171875,
-0.004535675048828125,
-0.05462646484375,
0.042510986328125,
-0.00894927978515625,
0.027191162109375,
-0.03607177734375,
-0.004550933837890625,
-0.0218658447265625,
-0.007110595703125,
-0.032745361328125,
-0.054351806640625,
-0.0019588470458984375,
-0.021697998046875,
-0.00844573974609375,
-0.02337646484375,
0.060089111328125,
-0.01708984375,
-0.04852294921875,
-0.0172119140625,
0.0255584716796875,
0.042816162109375,
-0.0171356201171875,
-0.052825927734375,
-0.003833770751953125,
-0.00266265869140625,
-0.019683837890625,
0.00508880615234375,
-0.01152801513671875,
0.0019989013671875,
0.0386962890625,
0.026275634765625,
-0.004253387451171875,
-0.023223876953125,
0.01457977294921875,
0.05426025390625,
-0.04071044921875,
-0.04571533203125,
-0.04644775390625,
0.032135009765625,
-0.01021575927734375,
-0.0361328125,
0.0467529296875,
0.03656005859375,
0.04986572265625,
-0.048126220703125,
0.02203369140625,
-0.0241546630859375,
0.017303466796875,
-0.01430511474609375,
0.045501708984375,
-0.04718017578125,
-0.0302276611328125,
-0.033416748046875,
-0.082763671875,
-0.018646240234375,
0.05316162109375,
0.004062652587890625,
0.0207061767578125,
0.0308837890625,
0.0679931640625,
0.0021152496337890625,
0.007427215576171875,
0.0028667449951171875,
-0.0171661376953125,
0.001956939697265625,
0.030517578125,
0.06396484375,
-0.02166748046875,
-0.007465362548828125,
-0.0318603515625,
-0.05010986328125,
-0.020751953125,
-0.061981201171875,
-0.053924560546875,
-0.045989990234375,
-0.052947998046875,
-0.036712646484375,
-0.002536773681640625,
0.08331298828125,
0.0814208984375,
-0.03839111328125,
-0.0299072265625,
0.0011005401611328125,
-0.0142059326171875,
-0.038726806640625,
-0.0196685791015625,
-0.00028014183044433594,
0.020660400390625,
-0.057952880859375,
-0.0014629364013671875,
0.01021575927734375,
0.04046630859375,
-0.016937255859375,
0.0006299018859863281,
0.0221405029296875,
-0.0288848876953125,
0.032470703125,
0.0521240234375,
-0.060394287109375,
-0.0211029052734375,
-0.03009033203125,
0.007152557373046875,
0.0199127197265625,
0.03936767578125,
-0.04705810546875,
0.03717041015625,
0.025115966796875,
0.016143798828125,
0.05181884765625,
0.015960693359375,
0.019744873046875,
-0.047454833984375,
0.004421234130859375,
0.01355743408203125,
0.0215911865234375,
0.017822265625,
-0.057525634765625,
0.03680419921875,
0.042083740234375,
-0.0239105224609375,
-0.0286712646484375,
0.0199432373046875,
-0.1126708984375,
0.0009293556213378906,
0.069091796875,
0.006809234619140625,
-0.01308441162109375,
0.01068115234375,
-0.033233642578125,
-0.01617431640625,
-0.037384033203125,
0.0531005859375,
0.052093505859375,
-0.045745849609375,
-0.02374267578125,
-0.05120849609375,
0.0199432373046875,
-0.004566192626953125,
-0.05682373046875,
-0.0148773193359375,
0.057861328125,
0.0240936279296875,
0.036407470703125,
0.055694580078125,
-0.0212554931640625,
0.040863037109375,
0.01148223876953125,
0.0284423828125,
-0.01381683349609375,
-0.01541900634765625,
-0.0185089111328125,
0.002727508544921875,
-0.01239013671875,
-0.036956787109375
]
] |
OpenBuddy/openbuddy-mistral-7b-v13 | 2023-10-11T15:54:14.000Z | [
"transformers",
"pytorch",
"mistral",
"text-generation",
"zh",
"en",
"fr",
"de",
"ja",
"ko",
"it",
"ru",
"license:apache-2.0",
"text-generation-inference",
"region:us"
] | text-generation | OpenBuddy | null | null | OpenBuddy/openbuddy-mistral-7b-v13 | 13 | 7,284 | transformers | 2023-10-10T06:48:00 | ---
language:
- zh
- en
- fr
- de
- ja
- ko
- it
- ru
pipeline_tag: text-generation
inference: false
library_name: transformers
license: apache-2.0
---
# OpenBuddy - Open Multilingual Chatbot
GitHub and Usage Guide: [https://github.com/OpenBuddy/OpenBuddy](https://github.com/OpenBuddy/OpenBuddy)
Website and Demo: [https://openbuddy.ai](https://openbuddy.ai)
Evaluation result of this model: [Evaluation.txt](Evaluation.txt)
# Copyright Notice
Base model: https://huggingface.co/mistralai/Mistral-7B-v0.1
License: Apache 2.0
## Disclaimer
All OpenBuddy models have inherent limitations and may potentially produce outputs that are erroneous, harmful, offensive, or otherwise undesirable. Users should not use these models in critical or high-stakes situations that may lead to personal injury, property damage, or significant losses. Examples of such scenarios include, but are not limited to, the medical field, controlling software and hardware systems that may cause harm, and making important financial or legal decisions.
OpenBuddy is provided "as-is" without any warranty of any kind, either express or implied, including, but not limited to, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement. In no event shall the authors, contributors, or copyright holders be liable for any claim, damages, or other liabilities, whether in an action of contract, tort, or otherwise, arising from, out of, or in connection with the software or the use or other dealings in the software.
By using OpenBuddy, you agree to these terms and conditions, and acknowledge that you understand the potential risks associated with its use. You also agree to indemnify and hold harmless the authors, contributors, and copyright holders from any claims, damages, or liabilities arising from your use of OpenBuddy.
## 免责声明
所有OpenBuddy模型均存在固有的局限性，可能产生错误的、有害的、冒犯性的或其他不良的输出。用户在关键或高风险场景中应谨慎行事，不要使用这些模型，以免导致人身伤害、财产损失或重大损失。此类场景的例子包括但不限于医疗领域、可能导致伤害的软硬件系统的控制以及进行重要的财务或法律决策。
OpenBuddy按“原样”提供，不附带任何种类的明示或暗示的保证，包括但不限于适销性、特定目的的适用性和非侵权的暗示保证。在任何情况下，作者、贡献者或版权所有者均不对因软件或使用或其他软件交易而产生的任何索赔、损害赔偿或其他责任（无论是合同、侵权还是其他原因）承担责任。
使用OpenBuddy即表示您同意这些条款和条件，并承认您了解其使用可能带来的潜在风险。您还同意赔偿并使作者、贡献者和版权所有者免受因您使用OpenBuddy而产生的任何索赔、损害赔偿或责任的影响。 | 2,330 | [
[
-0.0259857177734375,
-0.07427978515625,
0.01348114013671875,
0.038665771484375,
-0.0209503173828125,
-0.0130157470703125,
-0.017486572265625,
-0.033721923828125,
0.00962066650390625,
0.0295867919921875,
-0.0190277099609375,
-0.041748046875,
-0.033538818359375,
-0.019317626953125,
-0.002834320068359375,
0.0750732421875,
-0.0176239013671875,
-0.00499725341796875,
-0.00130462646484375,
-0.0159454345703125,
-0.049713134765625,
-0.0195159912109375,
-0.03857421875,
-0.00815582275390625,
0.003627777099609375,
0.027435302734375,
0.06427001953125,
-0.0012722015380859375,
0.0457763671875,
0.0281829833984375,
0.0034198760986328125,
-0.005706787109375,
-0.043365478515625,
0.0113983154296875,
0.00431060791015625,
-0.0309600830078125,
-0.049560546875,
-0.01192474365234375,
0.0141448974609375,
0.030029296875,
-0.02667236328125,
0.02703857421875,
0.004962921142578125,
0.051910400390625,
-0.05767822265625,
0.03082275390625,
-0.00787353515625,
0.00368499755859375,
-0.009063720703125,
-0.0224456787109375,
-0.01538848876953125,
-0.05810546875,
-0.00829315185546875,
-0.042694091796875,
-0.007762908935546875,
0.0088653564453125,
0.0823974609375,
-0.0011568069458007812,
-0.02337646484375,
-0.01444244384765625,
-0.057220458984375,
0.042816162109375,
-0.060272216796875,
0.0289154052734375,
0.02203369140625,
0.0545654296875,
-0.0194854736328125,
-0.04791259765625,
-0.039215087890625,
-0.0108795166015625,
-0.00305938720703125,
0.03021240234375,
-0.0251617431640625,
-0.0056610107421875,
0.01537322998046875,
0.041015625,
-0.056976318359375,
-0.006744384765625,
-0.04443359375,
-0.0037670135498046875,
0.0297393798828125,
0.0106964111328125,
0.045928955078125,
-0.0219573974609375,
-0.03863525390625,
-0.005329132080078125,
-0.02813720703125,
0.034942626953125,
0.0306396484375,
0.0202178955078125,
-0.05157470703125,
0.05743408203125,
-0.018768310546875,
0.032257080078125,
-0.00408935546875,
-0.028717041015625,
0.041473388671875,
-0.0325927734375,
-0.0279388427734375,
-0.0018053054809570312,
0.08148193359375,
0.045928955078125,
0.02569580078125,
0.007415771484375,
-0.00872039794921875,
-0.007671356201171875,
0.010589599609375,
-0.06353759765625,
-0.0213470458984375,
0.050079345703125,
-0.053131103515625,
-0.0199127197265625,
0.0106201171875,
-0.06744384765625,
-0.01207733154296875,
-0.0025463104248046875,
0.027374267578125,
-0.050323486328125,
-0.04974365234375,
0.0164337158203125,
-0.00609588623046875,
-0.0034503936767578125,
0.018218994140625,
-0.036529541015625,
0.01971435546875,
0.0152587890625,
0.0833740234375,
0.0224151611328125,
-0.01470947265625,
-0.005847930908203125,
0.019073486328125,
-0.0177459716796875,
0.04083251953125,
-0.01093292236328125,
-0.04083251953125,
0.0045928955078125,
0.00864410400390625,
0.002292633056640625,
-0.01485443115234375,
0.0258941650390625,
-0.0162811279296875,
0.0457763671875,
0.0265960693359375,
-0.01308441162109375,
-0.030914306640625,
0.0014286041259765625,
-0.040252685546875,
0.06988525390625,
0.0109100341796875,
-0.06976318359375,
0.0102386474609375,
-0.07366943359375,
-0.0280609130859375,
0.0020236968994140625,
-0.0126190185546875,
-0.032623291015625,
-0.002376556396484375,
0.01071929931640625,
0.033294677734375,
-0.0162200927734375,
0.0128936767578125,
-0.042999267578125,
-0.0158843994140625,
0.022125244140625,
-0.023529052734375,
0.10418701171875,
0.0184478759765625,
-0.0079345703125,
0.037261962890625,
-0.04498291015625,
0.01157379150390625,
0.038787841796875,
-0.0268096923828125,
-0.03216552734375,
-0.01739501953125,
0.0131683349609375,
0.0187835693359375,
0.026641845703125,
-0.0496826171875,
0.018218994140625,
-0.039947509765625,
0.034881591796875,
0.05474853515625,
0.0068206787109375,
0.0266571044921875,
-0.037322998046875,
0.054229736328125,
0.008209228515625,
0.039581298828125,
-0.02801513671875,
-0.057464599609375,
-0.038848876953125,
-0.042938232421875,
0.00757598876953125,
0.0584716796875,
-0.043701171875,
0.04705810546875,
-0.0139007568359375,
-0.054443359375,
-0.056854248046875,
-0.0048675537109375,
0.031219482421875,
0.02008056640625,
0.0265960693359375,
-0.01544952392578125,
-0.026611328125,
-0.041778564453125,
-0.005680084228515625,
-0.02581787109375,
-0.009613037109375,
0.03106689453125,
0.045379638671875,
-0.01052093505859375,
0.061767578125,
-0.056915283203125,
-0.034423828125,
0.00555419921875,
0.0020542144775390625,
0.0225677490234375,
0.053466796875,
0.06732177734375,
-0.05535888671875,
-0.049835205078125,
0.00637054443359375,
-0.066650390625,
0.01493072509765625,
-0.0018291473388671875,
-0.0264892578125,
0.027801513671875,
0.0183563232421875,
-0.059326171875,
0.07159423828125,
0.0484619140625,
-0.03521728515625,
0.056304931640625,
-0.029205322265625,
0.0189208984375,
-0.10333251953125,
0.01947021484375,
-0.01070404052734375,
-0.01432037353515625,
-0.036285400390625,
0.0218353271484375,
0.0084381103515625,
-0.01739501953125,
-0.039947509765625,
0.050079345703125,
-0.0268096923828125,
0.023834228515625,
-0.001827239990234375,
0.0173492431640625,
-0.0138397216796875,
0.035491943359375,
-0.0150909423828125,
0.04473876953125,
0.0419921875,
-0.031982421875,
0.0406494140625,
0.0272216796875,
-0.0255126953125,
0.04248046875,
-0.072509765625,
-0.01039886474609375,
-0.004680633544921875,
0.014984130859375,
-0.08514404296875,
-0.0243072509765625,
0.05328369140625,
-0.06903076171875,
0.0175933837890625,
-0.00696563720703125,
-0.043487548828125,
-0.033416748046875,
-0.0299072265625,
0.009124755859375,
0.045379638671875,
-0.025665283203125,
0.03057861328125,
0.0209197998046875,
-0.0207366943359375,
-0.042694091796875,
-0.049407958984375,
-0.0234832763671875,
-0.01058197021484375,
-0.06915283203125,
0.01238250732421875,
-0.0133819580078125,
-0.0020771026611328125,
0.00791168212890625,
0.008758544921875,
-0.0161895751953125,
-0.0044708251953125,
0.046295166015625,
0.02862548828125,
-0.01161956787109375,
0.006336212158203125,
0.003978729248046875,
-0.0079803466796875,
-0.0103912353515625,
0.004589080810546875,
0.04278564453125,
-0.020172119140625,
-0.038665771484375,
-0.0259552001953125,
0.037322998046875,
0.042205810546875,
-0.015838623046875,
0.062103271484375,
0.05682373046875,
-0.0367431640625,
0.008758544921875,
-0.032196044921875,
-0.0013284683227539062,
-0.03656005859375,
0.01483917236328125,
-0.03375244140625,
-0.06268310546875,
0.052825927734375,
0.0122222900390625,
0.02813720703125,
0.019378662109375,
0.057891845703125,
-0.0020732879638671875,
0.07305908203125,
0.0504150390625,
0.01380157470703125,
0.0300140380859375,
-0.01070404052734375,
0.0251922607421875,
-0.046112060546875,
-0.0259552001953125,
-0.047882080078125,
-0.01035308837890625,
-0.055206298828125,
-0.0245361328125,
0.0264129638671875,
0.0301361083984375,
-0.03997802734375,
0.02166748046875,
-0.0543212890625,
0.0238189697265625,
0.058502197265625,
0.0214691162109375,
0.0161285400390625,
-0.01061248779296875,
-0.0224456787109375,
0.015716552734375,
-0.038238525390625,
-0.040252685546875,
0.07366943359375,
0.0262603759765625,
0.06585693359375,
0.03228759765625,
0.04791259765625,
-0.0111846923828125,
0.01204681396484375,
-0.054840087890625,
0.0309600830078125,
0.0166778564453125,
-0.0716552734375,
-0.03594970703125,
-0.023529052734375,
-0.09600830078125,
0.02032470703125,
-0.0015096664428710938,
-0.07745361328125,
0.00829315185546875,
0.006336212158203125,
-0.017486572265625,
0.03363037109375,
-0.062744140625,
0.06658935546875,
-0.01355743408203125,
-0.0250091552734375,
-0.0067138671875,
-0.047393798828125,
0.040008544921875,
-0.002338409423828125,
0.0298309326171875,
-0.0193023681640625,
-0.005313873291015625,
0.030426025390625,
-0.047027587890625,
0.0657958984375,
-0.02001953125,
0.007476806640625,
0.02569580078125,
0.0283660888671875,
0.01074981689453125,
0.0182647705078125,
0.025115966796875,
0.042724609375,
0.0230560302734375,
-0.038238525390625,
-0.029266357421875,
0.056488037109375,
-0.0697021484375,
-0.03668212890625,
-0.038543701171875,
-0.02362060546875,
0.005924224853515625,
0.0362548828125,
0.0167999267578125,
0.0175018310546875,
-0.01361846923828125,
0.022857666015625,
0.00669097900390625,
-0.052734375,
0.031768798828125,
0.044647216796875,
-0.04205322265625,
-0.03924560546875,
0.06060791015625,
0.0002162456512451172,
0.01178741455078125,
0.0087890625,
0.01739501953125,
-0.014007568359375,
-0.027862548828125,
-0.0340576171875,
0.01873779296875,
-0.044677734375,
-0.023956298828125,
-0.030670166015625,
0.002902984619140625,
-0.05682373046875,
-0.01412200927734375,
-0.01375579833984375,
-0.0306549072265625,
-0.00566864013671875,
-0.00010842084884643555,
0.0439453125,
0.018341064453125,
-0.022674560546875,
0.0095672607421875,
-0.07989501953125,
0.0413818359375,
-0.004787445068359375,
0.057220458984375,
0.00115966796875,
-0.0175933837890625,
-0.0268096923828125,
0.01418304443359375,
-0.03704833984375,
-0.078369140625,
0.033203125,
-0.01983642578125,
0.048614501953125,
0.04583740234375,
0.0273284912109375,
0.050079345703125,
-0.029052734375,
0.060272216796875,
0.05853271484375,
-0.0504150390625,
0.05816650390625,
-0.0438232421875,
0.0276641845703125,
0.028778076171875,
0.060821533203125,
-0.040771484375,
-0.0250701904296875,
-0.043426513671875,
-0.060699462890625,
0.0655517578125,
0.0299835205078125,
0.006542205810546875,
0.0019989013671875,
-0.01092529296875,
0.002559661865234375,
0.022735595703125,
-0.058197021484375,
-0.0310516357421875,
-0.03607177734375,
-0.010772705078125,
0.0123443603515625,
0.0016574859619140625,
-0.018768310546875,
-0.01074981689453125,
0.050689697265625,
0.0109710693359375,
0.039093017578125,
0.00534820556640625,
0.005218505859375,
-0.0273590087890625,
0.021636962890625,
0.04351806640625,
0.05291748046875,
-0.041015625,
-0.0252532958984375,
-0.01459503173828125,
-0.035491943359375,
-0.003116607666015625,
0.016204833984375,
-0.01751708984375,
0.00324249267578125,
0.00966644287109375,
0.056732177734375,
0.01412200927734375,
-0.05224609375,
0.051605224609375,
-0.007720947265625,
0.00856781005859375,
-0.041473388671875,
-0.0042266845703125,
0.01520538330078125,
0.0196075439453125,
0.0016374588012695312,
0.0120849609375,
0.006389617919921875,
-0.039337158203125,
-0.01415252685546875,
0.0202178955078125,
-0.03753662109375,
-0.01500701904296875,
0.058990478515625,
0.0233612060546875,
-0.041229248046875,
0.04351806640625,
-0.0019130706787109375,
-0.01122283935546875,
0.043701171875,
0.022735595703125,
0.0731201171875,
-0.04412841796875,
0.00848388671875,
0.049713134765625,
0.029266357421875,
0.01739501953125,
0.053619384765625,
0.0102081298828125,
-0.044677734375,
-0.0296478271484375,
-0.0274200439453125,
-0.03424072265625,
0.015411376953125,
-0.059326171875,
0.0367431640625,
-0.036956787109375,
-0.0296173095703125,
0.006984710693359375,
-0.024871826171875,
-0.04559326171875,
-0.00806427001953125,
-0.0060577392578125,
0.069580078125,
-0.036376953125,
0.042999267578125,
0.06884765625,
-0.0709228515625,
-0.046295166015625,
-0.01491546630859375,
0.00455474853515625,
-0.051177978515625,
0.02801513671875,
0.015380859375,
0.00658416748046875,
-0.0292510986328125,
-0.036956787109375,
-0.05267333984375,
0.07354736328125,
0.0113983154296875,
-0.0199432373046875,
-0.01320648193359375,
-0.002685546875,
0.023712158203125,
0.0003159046173095703,
0.050689697265625,
-0.006938934326171875,
0.037994384765625,
-0.0106658935546875,
-0.10760498046875,
0.028961181640625,
-0.0277557373046875,
-0.008331298828125,
0.01377105712890625,
-0.06512451171875,
0.07440185546875,
-0.034210205078125,
-0.01019287109375,
0.00949859619140625,
0.03369140625,
0.0243988037109375,
0.028533935546875,
0.0306549072265625,
0.026458740234375,
0.035919189453125,
-0.01396942138671875,
0.07318115234375,
-0.03570556640625,
0.037200927734375,
0.06915283203125,
0.00615692138671875,
0.061981201171875,
0.01544189453125,
-0.03302001953125,
0.048919677734375,
0.037994384765625,
0.0016117095947265625,
0.0187530517578125,
0.0020732879638671875,
-0.004085540771484375,
-0.0019683837890625,
0.0089263916015625,
-0.04644775390625,
0.024688720703125,
0.0287933349609375,
-0.0192108154296875,
-0.0178985595703125,
0.01554107666015625,
0.0051727294921875,
-0.00966644287109375,
-0.005126953125,
0.05413818359375,
0.0010328292846679688,
-0.02935791015625,
0.055267333984375,
0.0086669921875,
0.0384521484375,
-0.061553955078125,
-0.004108428955078125,
-0.00757598876953125,
0.019287109375,
-0.02532958984375,
-0.059234619140625,
0.0037136077880859375,
-0.005214691162109375,
-0.0011339187622070312,
-0.0012149810791015625,
0.058197021484375,
-0.0054168701171875,
-0.01751708984375,
0.0264434814453125,
0.046630859375,
0.0205230712890625,
-0.00292205810546875,
-0.06671142578125,
-0.0009851455688476562,
-0.004108428955078125,
-0.042449951171875,
0.019012451171875,
0.039825439453125,
0.0039215087890625,
0.06683349609375,
0.053192138671875,
0.005161285400390625,
-0.004779815673828125,
-0.0021820068359375,
0.06939697265625,
-0.049285888671875,
-0.05755615234375,
-0.04095458984375,
0.06292724609375,
-0.0034027099609375,
-0.0284881591796875,
0.0648193359375,
0.052886962890625,
0.07257080078125,
-0.017547607421875,
0.0709228515625,
-0.0185394287109375,
0.051116943359375,
-0.019622802734375,
0.0577392578125,
-0.05078125,
-0.0277099609375,
-0.03533935546875,
-0.04705810546875,
-0.01168060302734375,
0.06231689453125,
-0.016387939453125,
0.0173492431640625,
0.04815673828125,
0.048187255859375,
-0.002834320068359375,
0.01187896728515625,
0.02001953125,
0.0297698974609375,
0.01282501220703125,
0.03985595703125,
0.04888916015625,
-0.03277587890625,
0.068115234375,
-0.028289794921875,
-0.038726806640625,
-0.0316162109375,
-0.034698486328125,
-0.0849609375,
-0.036224365234375,
-0.033172607421875,
-0.042449951171875,
-0.007610321044921875,
0.06781005859375,
0.05828857421875,
-0.066162109375,
-0.033111572265625,
0.01739501953125,
0.005496978759765625,
-0.03204345703125,
-0.02569580078125,
0.02392578125,
-0.006252288818359375,
-0.0693359375,
0.0022029876708984375,
0.01200103759765625,
0.01526641845703125,
-0.025238037109375,
0.0005011558532714844,
-0.0095977783203125,
0.0011138916015625,
0.046539306640625,
0.024566650390625,
-0.055572509765625,
-0.016815185546875,
-0.0086669921875,
0.0005655288696289062,
0.002246856689453125,
0.029876708984375,
-0.04150390625,
0.049407958984375,
0.051971435546875,
0.0032253265380859375,
0.0283966064453125,
-0.01314544677734375,
0.02001953125,
-0.037139892578125,
0.02508544921875,
0.007740020751953125,
0.038299560546875,
-0.0013990402221679688,
-0.0212554931640625,
0.04974365234375,
0.0103912353515625,
-0.0386962890625,
-0.06744384765625,
0.00605010986328125,
-0.07659912109375,
-0.037445068359375,
0.08062744140625,
-0.025421142578125,
0.0006198883056640625,
-0.00696563720703125,
-0.038330078125,
0.034454345703125,
-0.058197021484375,
0.051910400390625,
0.0418701171875,
-0.017791748046875,
-0.0013666152954101562,
-0.061492919921875,
0.007061004638671875,
-0.006572723388671875,
-0.057525634765625,
-0.0091705322265625,
0.048095703125,
0.0173492431640625,
0.0268096923828125,
0.058197021484375,
-0.0208892822265625,
0.0299072265625,
0.003459930419921875,
0.03436279296875,
-0.027313232421875,
-0.0006737709045410156,
-0.0114593505859375,
0.023345947265625,
-0.0240478515625,
-0.03662109375
]
] |
OpenBuddy/openbuddy-mistral-7b-v13-base | 2023-10-22T04:14:34.000Z | [
"transformers",
"pytorch",
"mistral",
"text-generation",
"zh",
"en",
"fr",
"de",
"ja",
"ko",
"it",
"ru",
"license:apache-2.0",
"text-generation-inference",
"region:us"
] | text-generation | OpenBuddy | null | null | OpenBuddy/openbuddy-mistral-7b-v13-base | 5 | 7,280 | transformers | 2023-10-11T06:56:16 | ---
language:
- zh
- en
- fr
- de
- ja
- ko
- it
- ru
pipeline_tag: text-generation
inference: false
library_name: transformers
license: apache-2.0
---
# ⚠️ About Base-series Models ⚠️
This is a part of the Base-series models, trained utilizing approximately 50% of conversational data. It embodies cognitive and dialogue capabilities parallel to the fully-trained OpenBuddy models, yet **it hasn’t been extensively fine-tuned for generic conversational tasks**.
We released this model intending to empower the community, enabling further fine-tuning and deployment of specialized, domain-specific models.
For immediate use in generic conversations, consider referring to our versions that without the `-base` suffix: https://huggingface.co/OpenBuddy/openbuddy-mistral-7b-v13.1
# OpenBuddy - Open Multilingual Chatbot
GitHub and Usage Guide: [https://github.com/OpenBuddy/OpenBuddy](https://github.com/OpenBuddy/OpenBuddy)
Website and Demo: [https://openbuddy.ai](https://openbuddy.ai)
# Copyright Notice
Base model: https://huggingface.co/mistralai/Mistral-7B-v0.1
License: Apache 2.0
## Disclaimer
All OpenBuddy models have inherent limitations and may potentially produce outputs that are erroneous, harmful, offensive, or otherwise undesirable. Users should not use these models in critical or high-stakes situations that may lead to personal injury, property damage, or significant losses. Examples of such scenarios include, but are not limited to, the medical field, controlling software and hardware systems that may cause harm, and making important financial or legal decisions.
OpenBuddy is provided "as-is" without any warranty of any kind, either express or implied, including, but not limited to, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement. In no event shall the authors, contributors, or copyright holders be liable for any claim, damages, or other liabilities, whether in an action of contract, tort, or otherwise, arising from, out of, or in connection with the software or the use or other dealings in the software.
By using OpenBuddy, you agree to these terms and conditions, and acknowledge that you understand the potential risks associated with its use. You also agree to indemnify and hold harmless the authors, contributors, and copyright holders from any claims, damages, or liabilities arising from your use of OpenBuddy.
## 免责声明
所有OpenBuddy模型均存在固有的局限性，可能产生错误的、有害的、冒犯性的或其他不良的输出。用户在关键或高风险场景中应谨慎行事，不要使用这些模型，以免导致人身伤害、财产损失或重大损失。此类场景的例子包括但不限于医疗领域、可能导致伤害的软硬件系统的控制以及进行重要的财务或法律决策。
OpenBuddy按“原样”提供，不附带任何种类的明示或暗示的保证，包括但不限于适销性、特定目的的适用性和非侵权的暗示保证。在任何情况下，作者、贡献者或版权所有者均不对因软件或使用或其他软件交易而产生的任何索赔、损害赔偿或其他责任（无论是合同、侵权还是其他原因）承担责任。
使用OpenBuddy即表示您同意这些条款和条件，并承认您了解其使用可能带来的潜在风险。您还同意赔偿并使作者、贡献者和版权所有者免受因您使用OpenBuddy而产生的任何索赔、损害赔偿或责任的影响。 | 2,893 | [
[
-0.0312042236328125,
-0.0689697265625,
0.0146942138671875,
0.02862548828125,
-0.025848388671875,
-0.0233154296875,
-0.01288604736328125,
-0.0285186767578125,
0.0016956329345703125,
0.037750244140625,
-0.032135009765625,
-0.042633056640625,
-0.03106689453125,
-0.0207366943359375,
-0.0006499290466308594,
0.0753173828125,
-0.0154571533203125,
-0.01381683349609375,
-0.0018644332885742188,
-0.0178375244140625,
-0.0458984375,
-0.0282440185546875,
-0.033447265625,
-0.0077972412109375,
-0.0030670166015625,
0.026763916015625,
0.06353759765625,
0.015838623046875,
0.042724609375,
0.0236663818359375,
0.00199127197265625,
-0.0005669593811035156,
-0.051025390625,
0.01204681396484375,
0.0034503936767578125,
-0.0299072265625,
-0.049072265625,
-0.01235198974609375,
0.02197265625,
0.048187255859375,
-0.0230865478515625,
0.032806396484375,
0.0002694129943847656,
0.058868408203125,
-0.055877685546875,
0.034759521484375,
-0.00957489013671875,
-0.00209808349609375,
-0.0103759765625,
-0.0187835693359375,
-0.019134521484375,
-0.053558349609375,
0.003368377685546875,
-0.046417236328125,
0.01093292236328125,
0.00849151611328125,
0.08599853515625,
0.006099700927734375,
-0.0236663818359375,
-0.005290985107421875,
-0.056884765625,
0.04193115234375,
-0.05706787109375,
0.025482177734375,
0.021514892578125,
0.042877197265625,
-0.017547607421875,
-0.052215576171875,
-0.032135009765625,
-0.005397796630859375,
-0.003864288330078125,
0.0212860107421875,
-0.0367431640625,
0.0017004013061523438,
0.0296478271484375,
0.050262451171875,
-0.05828857421875,
-0.0135955810546875,
-0.04547119140625,
0.005229949951171875,
0.027069091796875,
0.0203857421875,
0.029693603515625,
-0.01529693603515625,
-0.0379638671875,
-0.000652313232421875,
-0.039093017578125,
0.042205810546875,
0.017974853515625,
0.017242431640625,
-0.04150390625,
0.0556640625,
-0.027252197265625,
0.041015625,
0.01413726806640625,
-0.0196990966796875,
0.037872314453125,
-0.03302001953125,
-0.038055419921875,
0.0029010772705078125,
0.07061767578125,
0.041748046875,
0.0225067138671875,
0.0084381103515625,
-0.00591278076171875,
-0.00548553466796875,
0.01049041748046875,
-0.07373046875,
-0.01837158203125,
0.0462646484375,
-0.05047607421875,
-0.021392822265625,
0.010101318359375,
-0.053009033203125,
-0.003894805908203125,
-0.01446533203125,
0.027252197265625,
-0.049835205078125,
-0.050201416015625,
0.01678466796875,
-0.010833740234375,
-0.0012331008911132812,
0.0172271728515625,
-0.045135498046875,
0.034942626953125,
0.0212554931640625,
0.0731201171875,
0.01898193359375,
-0.00685882568359375,
-0.0060577392578125,
0.0277862548828125,
-0.0279083251953125,
0.037750244140625,
-0.01245880126953125,
-0.05108642578125,
0.0025844573974609375,
0.007312774658203125,
-0.0009937286376953125,
-0.0282440185546875,
0.01526641845703125,
-0.016357421875,
0.028717041015625,
0.007266998291015625,
-0.01739501953125,
-0.03131103515625,
-0.0026531219482421875,
-0.039642333984375,
0.07684326171875,
0.0216522216796875,
-0.066650390625,
0.0096588134765625,
-0.06396484375,
-0.02166748046875,
-0.0055999755859375,
-0.0088653564453125,
-0.0309906005859375,
0.00220489501953125,
0.0108184814453125,
0.026702880859375,
-0.0158233642578125,
0.01163482666015625,
-0.036712646484375,
-0.01325225830078125,
0.0162200927734375,
-0.0207977294921875,
0.09405517578125,
0.01245880126953125,
-0.01751708984375,
0.026885986328125,
-0.052276611328125,
0.01467132568359375,
0.0283660888671875,
-0.019805908203125,
-0.03533935546875,
-0.0236663818359375,
0.00630950927734375,
0.01354217529296875,
0.0251312255859375,
-0.04083251953125,
0.023223876953125,
-0.046142578125,
0.024871826171875,
0.05511474609375,
0.0169830322265625,
0.0299224853515625,
-0.037933349609375,
0.051483154296875,
0.02386474609375,
0.0292205810546875,
-0.03155517578125,
-0.0589599609375,
-0.045318603515625,
-0.04205322265625,
0.004665374755859375,
0.0538330078125,
-0.0309600830078125,
0.04901123046875,
-0.018402099609375,
-0.0494384765625,
-0.04315185546875,
-0.00445556640625,
0.039764404296875,
0.0333251953125,
0.028656005859375,
-0.024444580078125,
-0.02899169921875,
-0.05194091796875,
-0.0017337799072265625,
-0.0222625732421875,
0.005962371826171875,
0.0310211181640625,
0.039459228515625,
-0.006961822509765625,
0.081787109375,
-0.06781005859375,
-0.027374267578125,
0.0059356689453125,
0.00714874267578125,
0.02728271484375,
0.051727294921875,
0.07159423828125,
-0.046966552734375,
-0.04034423828125,
-0.0009555816650390625,
-0.06976318359375,
0.0094451904296875,
0.0045166015625,
-0.030303955078125,
0.0234832763671875,
0.02410888671875,
-0.0714111328125,
0.07305908203125,
0.03851318359375,
-0.033843994140625,
0.057281494140625,
-0.0222625732421875,
0.01271820068359375,
-0.10467529296875,
0.019805908203125,
-0.00135040283203125,
-0.007755279541015625,
-0.0458984375,
0.01453399658203125,
0.0031147003173828125,
-0.01666259765625,
-0.033355712890625,
0.04876708984375,
-0.0364990234375,
0.021148681640625,
-0.004241943359375,
-0.0026569366455078125,
-0.01290130615234375,
0.042572021484375,
-0.00904083251953125,
0.0406494140625,
0.045440673828125,
-0.04217529296875,
0.0249481201171875,
0.0245513916015625,
-0.02154541015625,
0.046630859375,
-0.064697265625,
-0.0116424560546875,
-0.01287078857421875,
0.0176544189453125,
-0.08966064453125,
-0.0239410400390625,
0.04681396484375,
-0.0595703125,
0.020751953125,
-0.0141754150390625,
-0.0280914306640625,
-0.0273895263671875,
-0.0279388427734375,
0.0190887451171875,
0.052581787109375,
-0.023529052734375,
0.04290771484375,
0.01203155517578125,
-0.02008056640625,
-0.04364013671875,
-0.045684814453125,
-0.033203125,
-0.00678253173828125,
-0.06951904296875,
0.01461029052734375,
-0.0142974853515625,
-0.01169586181640625,
0.018707275390625,
-0.00872802734375,
-0.0174407958984375,
-0.00986480712890625,
0.05963134765625,
0.02935791015625,
-0.0257110595703125,
-0.004215240478515625,
0.0135345458984375,
-0.009979248046875,
-0.00850677490234375,
-0.00225830078125,
0.03826904296875,
-0.0230865478515625,
-0.03741455078125,
-0.0187530517578125,
0.02825927734375,
0.04376220703125,
-0.01288604736328125,
0.05352783203125,
0.044708251953125,
-0.040008544921875,
0.0005321502685546875,
-0.038909912109375,
0.0004019737243652344,
-0.036285400390625,
0.013671875,
-0.03411865234375,
-0.060699462890625,
0.048004150390625,
0.01111602783203125,
0.029052734375,
0.027191162109375,
0.05938720703125,
0.0013303756713867188,
0.0650634765625,
0.0435791015625,
0.01153564453125,
0.029937744140625,
-0.02337646484375,
0.01471710205078125,
-0.05230712890625,
-0.022216796875,
-0.0438232421875,
-0.00506591796875,
-0.040618896484375,
-0.025177001953125,
0.0301513671875,
0.03009033203125,
-0.05078125,
0.018890380859375,
-0.0465087890625,
0.0174713134765625,
0.05718994140625,
0.0191192626953125,
0.020355224609375,
-0.0019235610961914062,
-0.0206756591796875,
0.013702392578125,
-0.026397705078125,
-0.0309600830078125,
0.079833984375,
0.02239990234375,
0.0648193359375,
0.03369140625,
0.047760009765625,
-0.00574493408203125,
0.0165557861328125,
-0.056182861328125,
0.033599853515625,
0.0172882080078125,
-0.06695556640625,
-0.026947021484375,
-0.0238189697265625,
-0.09002685546875,
0.02069091796875,
0.007110595703125,
-0.07568359375,
0.0242462158203125,
0.0118255615234375,
-0.031707763671875,
0.0265655517578125,
-0.0677490234375,
0.05999755859375,
-0.00540924072265625,
-0.02410888671875,
-0.00473785400390625,
-0.043701171875,
0.042510986328125,
0.00397491455078125,
0.0178070068359375,
-0.011444091796875,
-0.009429931640625,
0.0272064208984375,
-0.056610107421875,
0.06884765625,
-0.02301025390625,
0.01265716552734375,
0.03155517578125,
0.0216827392578125,
0.01056671142578125,
0.00946044921875,
0.0213623046875,
0.037109375,
0.018157958984375,
-0.03448486328125,
-0.032470703125,
0.055816650390625,
-0.06817626953125,
-0.0394287109375,
-0.04541015625,
-0.0205841064453125,
-0.004283905029296875,
0.031890869140625,
0.022064208984375,
0.03155517578125,
-0.01526641845703125,
0.02203369140625,
0.0257110595703125,
-0.032501220703125,
0.033966064453125,
0.0443115234375,
-0.033905029296875,
-0.050140380859375,
0.056396484375,
0.008087158203125,
0.021636962890625,
0.008148193359375,
0.011627197265625,
-0.0152130126953125,
-0.0290679931640625,
-0.0430908203125,
0.0224151611328125,
-0.045074462890625,
-0.0189208984375,
-0.032928466796875,
0.005222320556640625,
-0.05181884765625,
-0.00830078125,
-0.01204681396484375,
-0.035400390625,
-0.01311492919921875,
-0.0028553009033203125,
0.04193115234375,
0.0252685546875,
-0.0149993896484375,
0.0119171142578125,
-0.07318115234375,
0.042572021484375,
-0.003322601318359375,
0.05029296875,
0.00939178466796875,
-0.025787353515625,
-0.0268096923828125,
0.0106964111328125,
-0.037322998046875,
-0.072021484375,
0.03045654296875,
-0.0191802978515625,
0.047393798828125,
0.047698974609375,
0.0219573974609375,
0.05816650390625,
-0.030120849609375,
0.063720703125,
0.050506591796875,
-0.04949951171875,
0.0577392578125,
-0.049102783203125,
0.026153564453125,
0.03521728515625,
0.06024169921875,
-0.0263519287109375,
-0.0123748779296875,
-0.05377197265625,
-0.061920166015625,
0.0670166015625,
0.0289154052734375,
-0.00452423095703125,
0.004383087158203125,
-0.004909515380859375,
0.0007152557373046875,
0.0312042236328125,
-0.0462646484375,
-0.039398193359375,
-0.036346435546875,
-0.01483154296875,
0.01427459716796875,
0.005008697509765625,
-0.019500732421875,
-0.0205535888671875,
0.05450439453125,
0.01019287109375,
0.03411865234375,
0.01493072509765625,
0.0094451904296875,
-0.03802490234375,
0.01422119140625,
0.039154052734375,
0.0550537109375,
-0.048248291015625,
-0.031097412109375,
-0.0158538818359375,
-0.048919677734375,
0.000782012939453125,
0.0194091796875,
-0.0224151611328125,
0.007171630859375,
0.00008094310760498047,
0.06494140625,
0.0121002197265625,
-0.04290771484375,
0.043487548828125,
0.004184722900390625,
-0.0020351409912109375,
-0.04168701171875,
0.0007252693176269531,
0.01467132568359375,
0.01422882080078125,
0.009979248046875,
0.0265655517578125,
0.00862884521484375,
-0.047393798828125,
-0.01175689697265625,
0.018524169921875,
-0.03497314453125,
-0.0148162841796875,
0.051300048828125,
0.010284423828125,
-0.0234527587890625,
0.035736083984375,
0.005031585693359375,
-0.003299713134765625,
0.051910400390625,
0.0228424072265625,
0.0755615234375,
-0.034149169921875,
-0.00007963180541992188,
0.049407958984375,
0.021942138671875,
0.00408172607421875,
0.049468994140625,
0.012176513671875,
-0.03997802734375,
-0.0217437744140625,
-0.032135009765625,
-0.0357666015625,
0.01532745361328125,
-0.06707763671875,
0.035430908203125,
-0.053070068359375,
-0.034454345703125,
0.00344085693359375,
-0.023162841796875,
-0.047576904296875,
0.003299713134765625,
-0.0037078857421875,
0.07489013671875,
-0.0411376953125,
0.0413818359375,
0.06536865234375,
-0.0753173828125,
-0.05792236328125,
-0.00907135009765625,
0.0008349418640136719,
-0.043182373046875,
0.039276123046875,
0.01110076904296875,
0.0033473968505859375,
-0.0182342529296875,
-0.037933349609375,
-0.059051513671875,
0.075439453125,
0.01415252685546875,
-0.0211334228515625,
-0.018951416015625,
0.0013628005981445312,
0.0238037109375,
-0.0061492919921875,
0.0478515625,
0.00676727294921875,
0.03326416015625,
-0.00899505615234375,
-0.099853515625,
0.031768798828125,
-0.0286865234375,
-0.003467559814453125,
0.0153350830078125,
-0.062744140625,
0.08380126953125,
-0.032684326171875,
-0.0023975372314453125,
0.017822265625,
0.038116455078125,
0.033843994140625,
0.0292816162109375,
0.0308685302734375,
0.044464111328125,
0.041015625,
-0.0194091796875,
0.07672119140625,
-0.03375244140625,
0.03448486328125,
0.06988525390625,
0.001819610595703125,
0.0657958984375,
0.014190673828125,
-0.029693603515625,
0.056549072265625,
0.039703369140625,
0.0022678375244140625,
0.0289764404296875,
0.009796142578125,
-0.012176513671875,
-0.01007843017578125,
0.0096282958984375,
-0.0595703125,
0.01053619384765625,
0.03460693359375,
-0.02880859375,
-0.0128021240234375,
0.005908966064453125,
0.0032958984375,
-0.0186614990234375,
-0.01507568359375,
0.0518798828125,
0.0072479248046875,
-0.037384033203125,
0.063232421875,
0.001285552978515625,
0.0357666015625,
-0.06573486328125,
-0.00925445556640625,
-0.00675201416015625,
0.018951416015625,
-0.032928466796875,
-0.056427001953125,
-0.003330230712890625,
-0.0085601806640625,
-0.0182037353515625,
0.004852294921875,
0.048309326171875,
0.0016508102416992188,
-0.0220794677734375,
0.01422882080078125,
0.04425048828125,
0.0158843994140625,
-0.007114410400390625,
-0.04876708984375,
-0.007198333740234375,
-0.002490997314453125,
-0.04425048828125,
0.01459503173828125,
0.0433349609375,
-0.0030460357666015625,
0.056854248046875,
0.055908203125,
0.00470733642578125,
0.005096435546875,
-0.0004334449768066406,
0.07672119140625,
-0.05548095703125,
-0.053985595703125,
-0.040252685546875,
0.0640869140625,
-0.003620147705078125,
-0.031890869140625,
0.056182861328125,
0.05731201171875,
0.06915283203125,
-0.02386474609375,
0.08392333984375,
-0.0189666748046875,
0.050140380859375,
-0.020233154296875,
0.04278564453125,
-0.04486083984375,
-0.02154541015625,
-0.036895751953125,
-0.04931640625,
-0.0004143714904785156,
0.056884765625,
-0.0240936279296875,
0.0181121826171875,
0.050628662109375,
0.052398681640625,
-0.01047515869140625,
0.0152130126953125,
0.0094757080078125,
0.026702880859375,
0.00811004638671875,
0.04052734375,
0.042144775390625,
-0.03521728515625,
0.06298828125,
-0.03131103515625,
-0.041595458984375,
-0.0290985107421875,
-0.033203125,
-0.0860595703125,
-0.0289306640625,
-0.029998779296875,
-0.04718017578125,
-0.004970550537109375,
0.07635498046875,
0.060516357421875,
-0.06884765625,
-0.034881591796875,
0.011016845703125,
-0.0099029541015625,
-0.034912109375,
-0.0238189697265625,
0.02691650390625,
-0.006916046142578125,
-0.058563232421875,
0.0071258544921875,
0.00856781005859375,
0.00887298583984375,
-0.0237884521484375,
-0.002094268798828125,
-0.0037746429443359375,
0.0018787384033203125,
0.04302978515625,
0.029144287109375,
-0.05206298828125,
-0.0082244873046875,
-0.0021648406982421875,
-0.0030384063720703125,
0.00621795654296875,
0.03900146484375,
-0.0504150390625,
0.040069580078125,
0.0516357421875,
0.00787353515625,
0.03826904296875,
0.00595855712890625,
0.0211334228515625,
-0.038177490234375,
0.0175628662109375,
0.003757476806640625,
0.0416259765625,
-0.0079345703125,
-0.0276947021484375,
0.049285888671875,
0.02752685546875,
-0.02996826171875,
-0.06396484375,
0.00921630859375,
-0.0980224609375,
-0.0286407470703125,
0.0841064453125,
-0.02099609375,
-0.003108978271484375,
-0.0118255615234375,
-0.0367431640625,
0.0212860107421875,
-0.061187744140625,
0.0313720703125,
0.040252685546875,
-0.0286712646484375,
0.003971099853515625,
-0.059967041015625,
0.01317596435546875,
-0.004428863525390625,
-0.04718017578125,
-0.005859375,
0.053070068359375,
0.033782958984375,
0.032440185546875,
0.0660400390625,
-0.018524169921875,
0.03375244140625,
0.006134033203125,
0.026763916015625,
-0.03375244140625,
0.005519866943359375,
-0.01139068603515625,
0.0298004150390625,
-0.0171051025390625,
-0.0310211181640625
]
] |
anton-l/wav2vec2-base-superb-sv | 2022-11-11T19:30:49.000Z | [
"transformers",
"pytorch",
"wav2vec2",
"audio-xvector",
"speech",
"audio",
"audio-classification",
"en",
"dataset:superb",
"arxiv:2105.01051",
"arxiv:1910.09700",
"arxiv:2006.11477",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | audio-classification | anton-l | null | null | anton-l/wav2vec2-base-superb-sv | 1 | 7,277 | transformers | 2022-03-02T23:29:05 | ---
language: en
datasets:
- superb
tags:
- speech
- audio
- wav2vec2
- audio-classification
license: apache-2.0
---
# Model Card for wav2vec2-base-superb-sv
# Model Details
## Model Description
- **Developed by:** Shu-wen Yang et al.
- **Shared by:** Anton Lozhkov
- **Model type:** Wav2Vec2 with an XVector head
- **Language(s) (NLP):** English
- **License:** Apache 2.0
- **Related Models:**
- **Parent Model:** wav2vec2-large-lv60
- **Resources for more information:**
- [GitHub Repo](https://github.com/s3prl/s3prl/tree/master/s3prl/downstream/sv_voxceleb1)
- [Associated Paper](https://arxiv.org/abs/2105.010517)
# Uses
## Direct Use
This is a ported version of
[S3PRL's Wav2Vec2 for the SUPERB Speaker Verification task](https://github.com/s3prl/s3prl/tree/master/s3prl/downstream/sv_voxceleb1).
The base model is [wav2vec2-large-lv60](https://huggingface.co/facebook/wav2vec2-large-lv60), which is pretrained on 16kHz
sampled speech audio. When using the model make sure that your speech input is also sampled at 16Khz.
For more information refer to [SUPERB: Speech processing Universal PERformance Benchmark](https://arxiv.org/abs/2105.01051)
## Out-of-Scope Use
The model should not be used to intentionally create hostile or alienating environments for people.
# Bias, Risks, and Limitations
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
## Recommendations
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
# Training Details
## Training Data
See the [superb dataset card](https://huggingface.co/datasets/superb)
## Training Procedure
### Preprocessing
More information needed
### Speeds, Sizes, Times
More information needed
# Evaluation
## Testing Data, Factors & Metrics
### Testing Data
See the [superb dataset card](https://huggingface.co/datasets/superb)
### Factors
### Metrics
More information needed
## Results
More information needed
# Model Examination
More information needed
# Environmental Impact
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
**BibTeX:**
```
@misc{https://doi.org/10.48550/arxiv.2006.11477,
doi = {10.48550/ARXIV.2006.11477},
url = {https://arxiv.org/abs/2006.11477},
author = {Baevski, Alexei and Zhou, Henry and Mohamed, Abdelrahman and Auli, Michael},
keywords = {Computation and Language (cs.CL), Machine Learning (cs.LG), Sound (cs.SD), Audio and Speech Processing (eess.AS), FOS: Computer and information sciences, FOS: Computer and information sciences, FOS: Electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering},
title = {wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations},
publisher = {arXiv},
@misc{https://doi.org/10.48550/arxiv.2105.01051,
doi = {10.48550/ARXIV.2105.01051},
url = {https://arxiv.org/abs/2105.01051},
author = {Yang, Shu-wen and Chi, Po-Han and Chuang, Yung-Sung and Lai, Cheng-I Jeff and Lakhotia, Kushal and Lin, Yist Y. and Liu, Andy T. and Shi, Jiatong and Chang, Xuankai and Lin, Guan-Ting and Huang, Tzu-Hsien and Tseng, Wei-Cheng and Lee, Ko-tik and Liu, Da-Rong and Huang, Zili and Dong, Shuyan and Li, Shang-Wen and Watanabe, Shinji and Mohamed, Abdelrahman and Lee, Hung-yi},
keywords = {Computation and Language (cs.CL), Sound (cs.SD), Audio and Speech Processing (eess.AS), FOS: Computer and information sciences, FOS: Computer and information sciences, FOS: Electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering},
title = {SUPERB: Speech processing Universal PERformance Benchmark},
publisher = {arXiv},
year = {2021},
}
```
# Glossary [optional]
More information needed
# More Information [optional]
More information needed
# Model Card Authors [optional]
Anton Lozhkov in collaboration with Ezi Ozoani and the Hugging Face team
# Model Card Contact
More information needed
# How to Get Started with the Model
Use the code below to get started with the model.
<details>
<summary> Click to expand </summary>
```python
from transformers import AutoProcessor, AutoModelForAudioXVector
processor = AutoProcessor.from_pretrained("anton-l/wav2vec2-base-superb-sv")
model = AutoModelForAudioXVector.from_pretrained("anton-l/wav2vec2-base-superb-sv")
```
</details>
| 5,578 | [
[
-0.033538818359375,
-0.038665771484375,
0.0200958251953125,
0.007007598876953125,
-0.009368896484375,
-0.01849365234375,
-0.01389312744140625,
-0.03961181640625,
-0.01114654541015625,
0.034912109375,
-0.04888916015625,
-0.034576416015625,
-0.046661376953125,
-0.0236968994140625,
-0.0167388916015625,
0.05499267578125,
0.0307769775390625,
0.01947021484375,
-0.0178070068359375,
-0.0080108642578125,
-0.03167724609375,
-0.05328369140625,
-0.036956787109375,
-0.0355224609375,
-0.0143280029296875,
0.0160064697265625,
0.020416259765625,
0.04486083984375,
0.0157318115234375,
0.0267791748046875,
-0.0306549072265625,
0.0038433074951171875,
-0.035003662109375,
-0.00585174560546875,
-0.00897979736328125,
-0.028656005859375,
-0.03265380859375,
-0.00004649162292480469,
0.05072021484375,
0.036773681640625,
-0.0238800048828125,
0.0290069580078125,
0.01263427734375,
0.028076171875,
-0.02093505859375,
0.037353515625,
-0.0548095703125,
-0.0006871223449707031,
-0.0116424560546875,
-0.007503509521484375,
-0.019256591796875,
0.0029163360595703125,
-0.005100250244140625,
-0.036102294921875,
0.00838470458984375,
-0.0108642578125,
0.06683349609375,
0.028167724609375,
-0.01331329345703125,
-0.007442474365234375,
-0.05596923828125,
0.06744384765625,
-0.06512451171875,
0.054229736328125,
0.0312347412109375,
0.0063018798828125,
-0.0130462646484375,
-0.055084228515625,
-0.054107666015625,
-0.01142120361328125,
0.0255126953125,
0.0285491943359375,
-0.03741455078125,
-0.00428009033203125,
0.02740478515625,
0.0218658447265625,
-0.039642333984375,
0.025360107421875,
-0.0328369140625,
-0.038818359375,
0.053741455078125,
0.0010128021240234375,
0.0153045654296875,
-0.0265350341796875,
-0.0162353515625,
-0.029541015625,
-0.031585693359375,
0.023162841796875,
0.0303955078125,
0.032562255859375,
-0.043121337890625,
0.0316162109375,
-0.002269744873046875,
0.030670166015625,
0.0024929046630859375,
-0.0149993896484375,
0.05145263671875,
-0.035614013671875,
-0.0038127899169921875,
0.0021610260009765625,
0.07464599609375,
0.00888824462890625,
0.0034770965576171875,
0.019744873046875,
-0.0038604736328125,
0.002231597900390625,
-0.00135040283203125,
-0.0548095703125,
-0.0188446044921875,
0.02520751953125,
-0.0273590087890625,
-0.0021038055419921875,
-0.00434112548828125,
-0.039825439453125,
0.0003447532653808594,
-0.04180908203125,
0.036346435546875,
-0.02716064453125,
-0.037445068359375,
0.0009899139404296875,
-0.004985809326171875,
0.0251922607421875,
-0.00807952880859375,
-0.0677490234375,
0.03594970703125,
0.046234130859375,
0.04742431640625,
-0.0032215118408203125,
-0.007701873779296875,
-0.036102294921875,
-0.004161834716796875,
-0.0198822021484375,
0.045562744140625,
-0.0231781005859375,
-0.047210693359375,
-0.01059722900390625,
0.00024437904357910156,
0.011871337890625,
-0.03741455078125,
0.06549072265625,
-0.01390838623046875,
0.02178955078125,
-0.0100250244140625,
-0.048828125,
-0.0188446044921875,
-0.031341552734375,
-0.023040771484375,
0.09454345703125,
0.0031986236572265625,
-0.049285888671875,
-0.006053924560546875,
-0.03387451171875,
-0.033538818359375,
-0.00035452842712402344,
-0.022735595703125,
-0.045745849609375,
-0.006404876708984375,
0.0154876708984375,
0.037841796875,
-0.0255126953125,
0.020233154296875,
-0.01611328125,
-0.0177001953125,
-0.00330352783203125,
-0.02703857421875,
0.0828857421875,
0.0241241455078125,
-0.046661376953125,
0.0038394927978515625,
-0.07171630859375,
0.0058135986328125,
0.0139923095703125,
-0.01690673828125,
-0.000037610530853271484,
0.0004787445068359375,
0.01861572265625,
0.025604248046875,
0.0221099853515625,
-0.035614013671875,
-0.01120758056640625,
-0.039764404296875,
0.04388427734375,
0.054046630859375,
-0.01255035400390625,
0.0125579833984375,
-0.00324249267578125,
0.0140380859375,
-0.016998291015625,
0.0100250244140625,
0.0059661865234375,
-0.04473876953125,
-0.05963134765625,
-0.012969970703125,
0.02374267578125,
0.0528564453125,
-0.028167724609375,
0.0694580078125,
-0.01009368896484375,
-0.056976318359375,
-0.040374755859375,
-0.003871917724609375,
0.040313720703125,
0.03106689453125,
0.052459716796875,
-0.0152740478515625,
-0.06817626953125,
-0.0654296875,
0.0030384063720703125,
-0.0263824462890625,
-0.008056640625,
0.049041748046875,
0.029052734375,
-0.023345947265625,
0.06256103515625,
-0.029998779296875,
-0.033599853515625,
-0.0208587646484375,
0.00197601318359375,
0.0196685791015625,
0.0577392578125,
0.027984619140625,
-0.050537109375,
-0.008941650390625,
-0.0154266357421875,
-0.0418701171875,
-0.0125274658203125,
0.0042572021484375,
0.0045623779296875,
0.0266571044921875,
0.037109375,
-0.03338623046875,
0.01151275634765625,
0.045074462890625,
-0.0131072998046875,
0.052032470703125,
-0.0167083740234375,
-0.007366180419921875,
-0.082763671875,
0.0023860931396484375,
0.01465606689453125,
-0.00643157958984375,
-0.04150390625,
-0.0144500732421875,
-0.0116729736328125,
-0.0020618438720703125,
-0.043182373046875,
0.032135009765625,
-0.027191162109375,
-0.01244354248046875,
-0.0171661376953125,
-0.0007176399230957031,
-0.0145721435546875,
0.05169677734375,
0.01299285888671875,
0.05712890625,
0.066162109375,
-0.0513916015625,
0.0199737548828125,
0.0229339599609375,
-0.020111083984375,
0.030853271484375,
-0.064453125,
0.03375244140625,
0.0097808837890625,
0.028656005859375,
-0.07489013671875,
-0.003307342529296875,
0.01389312744140625,
-0.06866455078125,
0.042816162109375,
-0.0180511474609375,
-0.037261962890625,
-0.047821044921875,
0.0013265609741210938,
0.0215911865234375,
0.062103271484375,
-0.041900634765625,
0.03778076171875,
0.0657958984375,
-0.01242828369140625,
-0.02825927734375,
-0.063232421875,
-0.0125579833984375,
-0.0005555152893066406,
-0.038360595703125,
0.038177490234375,
-0.0148162841796875,
0.00882720947265625,
-0.01422882080078125,
-0.0153656005859375,
-0.0014438629150390625,
-0.01091766357421875,
0.03466796875,
0.023773193359375,
-0.003917694091796875,
-0.0012769699096679688,
-0.005527496337890625,
-0.02447509765625,
0.0137481689453125,
-0.0316162109375,
0.044219970703125,
-0.0157928466796875,
-0.00768280029296875,
-0.0709228515625,
0.02569580078125,
0.038330078125,
-0.0185089111328125,
0.0302581787109375,
0.060516357421875,
-0.04266357421875,
-0.006130218505859375,
-0.048736572265625,
-0.01294708251953125,
-0.038665771484375,
0.059844970703125,
-0.0245361328125,
-0.068115234375,
0.032073974609375,
0.0208282470703125,
-0.00830078125,
0.06591796875,
0.051025390625,
-0.0038127899169921875,
0.08770751953125,
0.037322998046875,
-0.01319122314453125,
0.02813720703125,
-0.0523681640625,
0.00689697265625,
-0.07489013671875,
-0.03515625,
-0.06304931640625,
0.01505279541015625,
-0.047027587890625,
-0.04937744140625,
0.0120086669921875,
0.004711151123046875,
-0.0243377685546875,
0.04400634765625,
-0.053619384765625,
0.0022068023681640625,
0.051605224609375,
-0.008392333984375,
-0.0118865966796875,
0.008575439453125,
-0.01715087890625,
-0.00693511962890625,
-0.04669189453125,
-0.0222930908203125,
0.0662841796875,
0.048736572265625,
0.033905029296875,
-0.000789642333984375,
0.042449951171875,
0.016876220703125,
-0.035400390625,
-0.051605224609375,
0.051116943359375,
-0.0282135009765625,
-0.0384521484375,
-0.026123046875,
-0.038238525390625,
-0.052093505859375,
0.016357421875,
-0.019805908203125,
-0.0689697265625,
0.0229644775390625,
0.0158538818359375,
-0.0275726318359375,
0.0210723876953125,
-0.042236328125,
0.048492431640625,
-0.00838470458984375,
-0.017608642578125,
-0.0297393798828125,
-0.04052734375,
0.004604339599609375,
0.01338958740234375,
0.0239105224609375,
-0.005222320556640625,
0.0262298583984375,
0.0853271484375,
-0.0293121337890625,
0.05908203125,
-0.03497314453125,
-0.002262115478515625,
0.0447998046875,
-0.016021728515625,
0.052581787109375,
-0.01045989990234375,
-0.01139068603515625,
0.054901123046875,
0.00687408447265625,
-0.01262664794921875,
-0.01678466796875,
0.0643310546875,
-0.0853271484375,
-0.0301666259765625,
-0.0169830322265625,
-0.0226898193359375,
-0.020782470703125,
0.00786590576171875,
0.041900634765625,
0.05224609375,
-0.0003256797790527344,
0.0189208984375,
0.048065185546875,
-0.0211029052734375,
0.019012451171875,
0.040557861328125,
0.009857177734375,
-0.043914794921875,
0.09014892578125,
0.03619384765625,
0.00928497314453125,
0.0134735107421875,
0.0218658447265625,
-0.051025390625,
-0.052642822265625,
-0.0137481689453125,
0.0199432373046875,
-0.043365478515625,
-0.0030975341796875,
-0.06927490234375,
-0.03387451171875,
-0.059417724609375,
0.0277862548828125,
-0.060791015625,
-0.033233642578125,
-0.047821044921875,
-0.007190704345703125,
0.018798828125,
0.034210205078125,
-0.038360595703125,
0.005550384521484375,
-0.0338134765625,
0.030670166015625,
0.0307464599609375,
0.01763916015625,
0.00223541259765625,
-0.0799560546875,
-0.01605224609375,
0.01160430908203125,
-0.0232086181640625,
-0.03851318359375,
0.0179901123046875,
0.01212310791015625,
0.06561279296875,
0.01751708984375,
-0.0008873939514160156,
0.042144775390625,
-0.036651611328125,
0.08526611328125,
0.0263214111328125,
-0.082275390625,
0.047393798828125,
-0.024810791015625,
0.0166473388671875,
0.0294342041015625,
0.0183868408203125,
-0.03369140625,
-0.01551055908203125,
-0.06591796875,
-0.06982421875,
0.07110595703125,
0.02825927734375,
0.0179901123046875,
0.006298065185546875,
0.017364501953125,
-0.0085601806640625,
-0.007678985595703125,
-0.04705810546875,
-0.05426025390625,
-0.033843994140625,
-0.004261016845703125,
-0.01134490966796875,
-0.03460693359375,
0.008575439453125,
-0.04351806640625,
0.0855712890625,
0.0036716461181640625,
0.04937744140625,
0.0199432373046875,
0.00836181640625,
0.00850677490234375,
0.0166168212890625,
0.0491943359375,
0.013641357421875,
-0.01715087890625,
0.00487518310546875,
0.0280303955078125,
-0.037811279296875,
-0.0003726482391357422,
0.0254364013671875,
0.005615234375,
-0.011474609375,
0.021575927734375,
0.0826416015625,
0.007080078125,
-0.02825927734375,
0.049407958984375,
0.001636505126953125,
-0.04083251953125,
-0.034637451171875,
0.0186767578125,
0.01922607421875,
0.01271820068359375,
0.029388427734375,
0.00807952880859375,
0.0244140625,
-0.028656005859375,
0.01351165771484375,
0.0281524658203125,
-0.05792236328125,
-0.0211029052734375,
0.06390380859375,
0.0228271484375,
-0.028564453125,
0.039031982421875,
-0.019317626953125,
-0.03955078125,
0.052459716796875,
0.0523681640625,
0.06719970703125,
-0.0266265869140625,
-0.00809478759765625,
0.053314208984375,
0.009613037109375,
-0.0012454986572265625,
0.02294921875,
-0.0169677734375,
-0.043670654296875,
-0.0272216796875,
-0.04705810546875,
-0.01898193359375,
0.0350341796875,
-0.047271728515625,
0.015960693359375,
-0.0297088623046875,
-0.0179901123046875,
0.005825042724609375,
0.017669677734375,
-0.05828857421875,
0.0190887451171875,
0.0450439453125,
0.062103271484375,
-0.0576171875,
0.07110595703125,
0.03179931640625,
-0.0196380615234375,
-0.08380126953125,
-0.0186309814453125,
0.01824951171875,
-0.05084228515625,
0.029052734375,
0.01026153564453125,
-0.027923583984375,
0.0083465576171875,
-0.042510986328125,
-0.07379150390625,
0.07659912109375,
0.02825927734375,
-0.06707763671875,
0.0265350341796875,
0.0013580322265625,
0.03851318359375,
-0.0224456787109375,
0.02850341796875,
0.0440673828125,
0.02813720703125,
0.01250457763671875,
-0.08258056640625,
-0.00539398193359375,
-0.032745361328125,
0.00604248046875,
-0.0232086181640625,
-0.05975341796875,
0.048492431640625,
-0.02789306640625,
-0.0199737548828125,
0.0030460357666015625,
0.067138671875,
0.030242919921875,
0.01007843017578125,
0.04345703125,
0.040374755859375,
0.0643310546875,
-0.0054473876953125,
0.061126708984375,
-0.0239715576171875,
0.04150390625,
0.098876953125,
0.0015850067138671875,
0.0709228515625,
0.01508331298828125,
-0.045135498046875,
0.0325927734375,
0.035400390625,
-0.01491546630859375,
0.04425048828125,
0.0175018310546875,
-0.0070648193359375,
-0.029052734375,
-0.017730712890625,
-0.0638427734375,
0.04681396484375,
0.0202789306640625,
-0.01605224609375,
0.0210723876953125,
-0.0010251998901367188,
-0.00527191162109375,
-0.003589630126953125,
-0.0199127197265625,
0.0528564453125,
0.01316070556640625,
-0.0280303955078125,
0.060699462890625,
0.002414703369140625,
0.057525634765625,
-0.043365478515625,
-0.00489044189453125,
0.012969970703125,
0.0030384063720703125,
-0.0236358642578125,
-0.0565185546875,
0.006572723388671875,
-0.00872802734375,
-0.03271484375,
0.00673675537109375,
0.0401611328125,
-0.023162841796875,
-0.03948974609375,
0.044647216796875,
0.004734039306640625,
0.026641845703125,
0.00392913818359375,
-0.06658935546875,
0.0307464599609375,
0.019500732421875,
-0.01788330078125,
0.0013246536254882812,
-0.001689910888671875,
0.0211181640625,
0.03375244140625,
0.060211181640625,
0.010955810546875,
-0.00565338134765625,
0.024871826171875,
0.048492431640625,
-0.041015625,
-0.06134033203125,
-0.042633056640625,
0.041290283203125,
-0.00734710693359375,
-0.0219268798828125,
0.0428466796875,
0.050750732421875,
0.062744140625,
0.011199951171875,
0.07757568359375,
0.004322052001953125,
0.05224609375,
-0.036163330078125,
0.0478515625,
-0.036651611328125,
0.0220794677734375,
-0.0300140380859375,
-0.061309814453125,
-0.0102386474609375,
0.057861328125,
-0.0157623291015625,
0.01180267333984375,
0.0229644775390625,
0.0728759765625,
0.007541656494140625,
0.00986480712890625,
0.0247955322265625,
0.04986572265625,
0.033660888671875,
0.0316162109375,
0.040771484375,
-0.045135498046875,
0.0440673828125,
-0.0308074951171875,
-0.02764892578125,
-0.00556182861328125,
-0.0445556640625,
-0.046875,
-0.063232421875,
-0.044830322265625,
-0.0308074951171875,
0.004764556884765625,
0.0782470703125,
0.08355712890625,
-0.0662841796875,
-0.0203399658203125,
-0.0104217529296875,
-0.00783538818359375,
-0.0233306884765625,
-0.0165252685546875,
0.0416259765625,
-0.0018672943115234375,
-0.0491943359375,
0.0465087890625,
0.005374908447265625,
0.01033782958984375,
-0.021331787109375,
-0.0289306640625,
0.00016164779663085938,
0.00582122802734375,
0.036163330078125,
0.0311126708984375,
-0.06610107421875,
-0.0208587646484375,
-0.0230865478515625,
-0.0026035308837890625,
0.0125579833984375,
0.0240936279296875,
-0.054595947265625,
0.035552978515625,
0.0330810546875,
0.0188751220703125,
0.0526123046875,
-0.0256805419921875,
0.0194091796875,
-0.0303192138671875,
0.003826141357421875,
0.007099151611328125,
0.034637451171875,
0.03265380859375,
-0.030670166015625,
0.0159759521484375,
0.03204345703125,
-0.052093505859375,
-0.061431884765625,
-0.00514984130859375,
-0.09918212890625,
-0.0195159912109375,
0.1085205078125,
0.0106353759765625,
-0.0046234130859375,
-0.0081024169921875,
-0.0305023193359375,
0.043121337890625,
-0.03912353515625,
0.026641845703125,
0.0299224853515625,
-0.01015472412109375,
-0.004627227783203125,
-0.038970947265625,
0.03399658203125,
0.016357421875,
-0.04046630859375,
0.00908660888671875,
0.0307464599609375,
0.033538818359375,
0.0155487060546875,
0.04962158203125,
-0.0164794921875,
0.024505615234375,
0.00830078125,
0.038238525390625,
-0.02032470703125,
-0.0189208984375,
-0.03863525390625,
0.0029811859130859375,
-0.014923095703125,
-0.018707275390625
]
] |
CHIH-HUNG/llama-2-13b-FINETUNE5_4w-r4-q_k_v_o | 2023-10-07T06:15:59.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"dataset:huangyt/FINETUNE5",
"license:llama2",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | CHIH-HUNG | null | null | CHIH-HUNG/llama-2-13b-FINETUNE5_4w-r4-q_k_v_o | 0 | 7,277 | transformers | 2023-10-01T23:45:51 | ---
license: llama2
datasets:
- huangyt/FINETUNE5
---
# Model Card for Model ID
<!-- Provide a quick summary of what the model is/does. -->
在llama-2-13b上使用huangyt/FINETUNE5資料集進行訓練，總資料筆數約4w
# Fine-Tuning Information
- **GPU:** RTX4090 (single core / 24564MiB)
- **model:** meta-llama/Llama-2-13b-hf
- **dataset:** huangyt/FINETUNE3 (共約3.3w筆訓練集)
- **peft_type:** LoRA
- **lora_rank:** 16
- **lora_target:** q_proj, k_proj, v_proj, o_proj
- **per_device_train_batch_size:** 8
- **gradient_accumulation_steps:** 8
- **learning_rate :** 4e-4
- **epoch:** 1
- **precision:** bf16
- **quantization:** load_in_4bit
# Fine-Tuning Detail
- **train_loss:** 0.579
- **train_runtime:** 4:6:11 (use deepspeed)
# Evaluation
- 與Llama-2-13b比較4種Benchmark，包含**ARC**、**HellaSwag**、**MMLU**、**TruthfulQA**
- 評估結果使用**本地**所測的分數，並使用load_in_8bit
| Model | Average | ARC |HellaSwag| MMLU | TruthfulQA | Time (s) |
|---------------------------------------|---------|---------|---------|-------|------------|----------|
| FINETUNE5_4w-r4-q_k_v_o | 56.09 | 54.35 | 79.24 | 54.01 | 36.75 | 22095 |
| FINETUNE5_4w-r8-q_k_v_o | 57.55 | 55.38 | 79.57 | 54.03 | 41.21 | 22127 |
| FINETUNE5_4w-r16-q_k_v_o | 57.26 | 54.35 | 79.74 | 52.29 | 42.68 | 22153 |
| FINETUNE5_4w-r4-gate_up_down | 56.51 | 52.82 | 79.13 | 52.83 | 41.28 | 22899 |
| FINETUNE5_4w-r8-gate_up_down | 56.10 | 52.73 | 79.14 | 52.56 | 39.99 | 22926 |
| FINETUNE5_4w-r16-gate_up_down | 56.23 | 52.39 | 79.48 | 53.42 | 39.62 | 22963 |
| FINETUNE5_4w-r4-q_k_v_o_gate_up_down | 56.06 | 52.56 | 79.21 | 51.67 | 40.80 | 24303 |
| FINETUNE5_4w-r8-q_k_v_o_gate_up_down | 56.35 | 51.88 | 79.42 | 52.00 | 42.10 | 24376 |
| FINETUNE5_4w-r16-q_k_v_o_gate_up_down | 56.73 | 54.18 | 79.53 | 52.77 | 40.46 | 24439 |
# 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/FINETUNE5", 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 = "FINETUNE5.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}")
``` | 2,769 | [
[
-0.0521240234375,
-0.044036865234375,
0.01261138916015625,
0.008697509765625,
-0.039581298828125,
0.0022335052490234375,
-0.0139007568359375,
-0.01433563232421875,
0.014190673828125,
0.0360107421875,
-0.04962158203125,
-0.036590576171875,
-0.039947509765625,
0.008453369140625,
-0.01384735107421875,
0.07122802734375,
-0.00803375244140625,
-0.006938934326171875,
0.0263214111328125,
0.0013074874877929688,
-0.04083251953125,
-0.01290130615234375,
-0.0528564453125,
-0.0255889892578125,
0.0142059326171875,
0.020782470703125,
0.04815673828125,
0.0677490234375,
0.053375244140625,
0.02117919921875,
-0.011016845703125,
0.00888824462890625,
-0.040069580078125,
-0.0247650146484375,
0.02239990234375,
-0.0389404296875,
-0.0430908203125,
-0.0074310302734375,
0.0499267578125,
0.0296478271484375,
0.0031280517578125,
0.039794921875,
0.0106964111328125,
0.056060791015625,
-0.0254058837890625,
0.015899658203125,
-0.021942138671875,
0.010650634765625,
-0.026123046875,
-0.0299835205078125,
-0.00417327880859375,
-0.0257110595703125,
-0.010406494140625,
-0.0655517578125,
0.012481689453125,
0.01399993896484375,
0.1075439453125,
0.032196044921875,
-0.0198974609375,
0.003261566162109375,
-0.050994873046875,
0.06500244140625,
-0.076171875,
0.01056671142578125,
0.020416259765625,
0.0355224609375,
-0.003971099853515625,
-0.04681396484375,
-0.050872802734375,
0.01412200927734375,
-0.01068878173828125,
0.01922607421875,
-0.005176544189453125,
-0.01337432861328125,
0.035125732421875,
0.040679931640625,
-0.0377197265625,
-0.0018825531005859375,
-0.045654296875,
0.0017137527465820312,
0.06591796875,
0.031005859375,
0.00782012939453125,
-0.0298614501953125,
-0.023895263671875,
-0.016754150390625,
-0.036865234375,
0.02471923828125,
0.037841796875,
0.0299835205078125,
-0.038238525390625,
0.0318603515625,
-0.036651611328125,
0.03759765625,
0.01715087890625,
-0.023712158203125,
0.050750732421875,
-0.020172119140625,
-0.04522705078125,
0.006381988525390625,
0.07916259765625,
0.0421142578125,
-0.00789642333984375,
0.0169677734375,
-0.0111846923828125,
-0.01282501220703125,
-0.0031871795654296875,
-0.06610107421875,
-0.030242919921875,
0.042449951171875,
-0.052490234375,
-0.023345947265625,
0.00930023193359375,
-0.0672607421875,
0.00653076171875,
-0.0204620361328125,
0.0297393798828125,
-0.03192138671875,
-0.040618896484375,
0.0023822784423828125,
-0.0161590576171875,
0.0204315185546875,
0.0195159912109375,
-0.06463623046875,
0.0085601806640625,
0.0391845703125,
0.048828125,
0.0129547119140625,
-0.01751708984375,
-0.0061798095703125,
0.01427459716796875,
-0.025970458984375,
0.046905517578125,
0.004486083984375,
-0.026153564453125,
-0.0145721435546875,
0.02276611328125,
-0.005130767822265625,
-0.03973388671875,
0.057098388671875,
-0.03509521484375,
-0.0014085769653320312,
-0.033660888671875,
-0.017578125,
-0.0318603515625,
0.027923583984375,
-0.05377197265625,
0.0823974609375,
0.0111846923828125,
-0.06427001953125,
0.0293426513671875,
-0.046051025390625,
-0.022613525390625,
0.0067901611328125,
-0.0007801055908203125,
-0.0391845703125,
-0.0150604248046875,
0.0160980224609375,
0.038177490234375,
-0.0303192138671875,
0.01323699951171875,
-0.014801025390625,
-0.04541015625,
0.0121307373046875,
-0.033721923828125,
0.077880859375,
0.03997802734375,
-0.0245819091796875,
0.006587982177734375,
-0.07940673828125,
0.00876617431640625,
0.03741455078125,
-0.038848876953125,
-0.00693511962890625,
-0.01103973388671875,
0.0029201507568359375,
0.002582550048828125,
0.0235137939453125,
-0.01739501953125,
0.01389312744140625,
-0.0256195068359375,
0.03765869140625,
0.06182861328125,
-0.003971099853515625,
0.01367950439453125,
-0.0338134765625,
0.0279083251953125,
0.01153564453125,
0.0176849365234375,
-0.0023059844970703125,
-0.027496337890625,
-0.0679931640625,
-0.01541900634765625,
0.0130462646484375,
0.03857421875,
-0.03936767578125,
0.049774169921875,
-0.024139404296875,
-0.046600341796875,
-0.054779052734375,
-0.0012140274047851562,
0.0169219970703125,
0.035980224609375,
0.04052734375,
0.012054443359375,
-0.056427001953125,
-0.06884765625,
0.00018668174743652344,
0.006397247314453125,
0.0052642822265625,
0.0291900634765625,
0.045745849609375,
-0.0120391845703125,
0.04351806640625,
-0.041748046875,
-0.022918701171875,
-0.027008056640625,
0.0030612945556640625,
0.07220458984375,
0.044586181640625,
0.0511474609375,
-0.040191650390625,
-0.04718017578125,
0.00717926025390625,
-0.07623291015625,
0.00982666015625,
-0.0099639892578125,
-0.01160430908203125,
-0.00801849365234375,
0.01003265380859375,
-0.04132080078125,
0.0322265625,
0.03497314453125,
-0.0185699462890625,
0.052001953125,
-0.00241851806640625,
0.030029296875,
-0.0797119140625,
0.01558685302734375,
-0.0156707763671875,
0.0009026527404785156,
-0.0307159423828125,
0.00965118408203125,
-0.006450653076171875,
0.010833740234375,
-0.03302001953125,
0.018310546875,
-0.03717041015625,
0.00787353515625,
-0.007633209228515625,
-0.00560760498046875,
-0.0006575584411621094,
0.051605224609375,
-0.0169219970703125,
0.055145263671875,
0.038116455078125,
-0.062042236328125,
0.03997802734375,
0.02197265625,
-0.038116455078125,
0.0160369873046875,
-0.038238525390625,
-0.0043182373046875,
-0.0024566650390625,
0.011322021484375,
-0.07550048828125,
-0.0261077880859375,
0.0369873046875,
-0.034210205078125,
0.0204010009765625,
-0.028076171875,
-0.0187225341796875,
-0.052093505859375,
-0.0251922607421875,
0.019805908203125,
0.0237884521484375,
-0.039276123046875,
0.0212860107421875,
0.01372528076171875,
0.01263427734375,
-0.042755126953125,
-0.0635986328125,
-0.01525115966796875,
-0.021728515625,
-0.037811279296875,
0.026611328125,
-0.0045623779296875,
-0.003154754638671875,
0.01001739501953125,
-0.00443267822265625,
-0.003589630126953125,
0.00791168212890625,
0.01042938232421875,
0.036224365234375,
-0.0297393798828125,
-0.032318115234375,
0.011688232421875,
-0.0127105712890625,
0.00933837890625,
0.0096893310546875,
0.06341552734375,
-0.01024627685546875,
-0.008758544921875,
-0.05853271484375,
0.01139068603515625,
0.0321044921875,
-0.00286865234375,
0.040130615234375,
0.06048583984375,
-0.0178070068359375,
0.00013494491577148438,
-0.0159912109375,
0.00015246868133544922,
-0.038360595703125,
0.0248870849609375,
-0.0472412109375,
-0.04913330078125,
0.051910400390625,
-0.006496429443359375,
0.01363372802734375,
0.069091796875,
0.02215576171875,
-0.0203704833984375,
0.08465576171875,
0.0010232925415039062,
-0.0097503662109375,
0.0172271728515625,
-0.0760498046875,
0.006351470947265625,
-0.07159423828125,
-0.025634765625,
-0.041748046875,
-0.041595458984375,
-0.047576904296875,
-0.00860595703125,
0.0199432373046875,
0.022247314453125,
-0.04931640625,
0.0307464599609375,
-0.0635986328125,
0.017913818359375,
0.0469970703125,
0.0108642578125,
0.0094757080078125,
-0.00981903076171875,
0.00470733642578125,
-0.0010404586791992188,
-0.041412353515625,
-0.0224456787109375,
0.0933837890625,
0.0213775634765625,
0.049774169921875,
0.00832366943359375,
0.048095703125,
0.011688232421875,
-0.00533294677734375,
-0.05023193359375,
0.03875732421875,
-0.004093170166015625,
-0.05963134765625,
-0.01515960693359375,
-0.0281219482421875,
-0.053375244140625,
0.0231475830078125,
-0.02215576171875,
-0.056427001953125,
0.01041412353515625,
0.006954193115234375,
-0.04327392578125,
0.044158935546875,
-0.037261962890625,
0.05633544921875,
-0.026458740234375,
-0.0234527587890625,
0.004909515380859375,
-0.04498291015625,
0.051422119140625,
0.01476287841796875,
0.017425537109375,
-0.021575927734375,
0.02685546875,
0.07470703125,
-0.049468994140625,
0.02874755859375,
-0.0230865478515625,
0.0028934478759765625,
0.039703369140625,
0.001483917236328125,
0.04962158203125,
0.0208892822265625,
0.003753662109375,
0.031982421875,
0.00873565673828125,
-0.0161285400390625,
-0.0273895263671875,
0.06268310546875,
-0.0849609375,
-0.038238525390625,
-0.0426025390625,
-0.029266357421875,
0.01552581787109375,
0.032958984375,
0.04510498046875,
0.01323699951171875,
0.018707275390625,
0.02197265625,
0.041229248046875,
-0.0016202926635742188,
0.05023193359375,
0.015838623046875,
-0.01727294921875,
-0.056671142578125,
0.064697265625,
0.00760650634765625,
0.005901336669921875,
0.026214599609375,
0.01175689697265625,
-0.0284423828125,
-0.042755126953125,
-0.045501708984375,
0.01194000244140625,
-0.032989501953125,
-0.038116455078125,
-0.032684326171875,
-0.033233642578125,
-0.049102783203125,
-0.00136566162109375,
-0.044036865234375,
-0.0179443359375,
-0.0452880859375,
0.0000743865966796875,
0.043121337890625,
0.02935791015625,
-0.004009246826171875,
0.050445556640625,
-0.065673828125,
0.029052734375,
0.00688934326171875,
0.01947021484375,
0.00888824462890625,
-0.055084228515625,
-0.0293121337890625,
0.00389862060546875,
-0.0361328125,
-0.046600341796875,
0.044219970703125,
-0.00698089599609375,
0.03436279296875,
0.059478759765625,
-0.0022411346435546875,
0.0843505859375,
-0.022247314453125,
0.06854248046875,
0.0206298828125,
-0.0543212890625,
0.042083740234375,
-0.0362548828125,
0.001422882080078125,
0.04547119140625,
0.0248870849609375,
-0.029815673828125,
0.008087158203125,
-0.044586181640625,
-0.061004638671875,
0.0849609375,
0.01378631591796875,
-0.0154571533203125,
0.0228118896484375,
0.013427734375,
0.00296783447265625,
0.0187530517578125,
-0.054779052734375,
-0.052398681640625,
-0.043121337890625,
-0.00661468505859375,
0.0009174346923828125,
-0.000560760498046875,
-0.0204925537109375,
-0.03704833984375,
0.059478759765625,
0.0032978057861328125,
0.0430908203125,
0.0177459716796875,
0.01444244384765625,
-0.016265869140625,
0.0145263671875,
0.032135009765625,
0.030242919921875,
-0.049652099609375,
-0.00698089599609375,
0.007965087890625,
-0.039276123046875,
0.0020084381103515625,
0.0105133056640625,
-0.0235137939453125,
-0.0018672943115234375,
0.0250396728515625,
0.06927490234375,
0.0038166046142578125,
-0.0190887451171875,
0.020294189453125,
0.007129669189453125,
-0.02752685546875,
-0.036651611328125,
0.0171051025390625,
-0.00470733642578125,
0.0305633544921875,
0.0391845703125,
0.0082550048828125,
0.01438140869140625,
-0.024658203125,
-0.00971221923828125,
0.0201416015625,
-0.0036163330078125,
-0.017059326171875,
0.07342529296875,
0.005748748779296875,
-0.016021728515625,
0.040130615234375,
-0.01401519775390625,
-0.033660888671875,
0.057769775390625,
0.03851318359375,
0.056427001953125,
-0.0128936767578125,
-0.00606536865234375,
0.0643310546875,
0.0266876220703125,
-0.01007080078125,
0.045623779296875,
0.00223541259765625,
-0.047149658203125,
-0.0173797607421875,
-0.0487060546875,
-0.0011243820190429688,
0.047943115234375,
-0.06341552734375,
0.024139404296875,
-0.052459716796875,
-0.02618408203125,
-0.0029315948486328125,
0.0238800048828125,
-0.057098388671875,
0.033447265625,
0.00681304931640625,
0.0667724609375,
-0.05963134765625,
0.0670166015625,
0.030242919921875,
-0.04150390625,
-0.086669921875,
-0.0240478515625,
-0.01502227783203125,
-0.0609130859375,
0.037353515625,
0.01015472412109375,
0.021331787109375,
0.00739288330078125,
-0.0556640625,
-0.07720947265625,
0.10552978515625,
0.004299163818359375,
-0.048126220703125,
0.0036640167236328125,
0.0167388916015625,
0.0190582275390625,
-0.009124755859375,
0.030242919921875,
0.056884765625,
0.045654296875,
0.0028553009033203125,
-0.052642822265625,
0.0194244384765625,
-0.0307159423828125,
-0.007488250732421875,
0.019805908203125,
-0.08502197265625,
0.09869384765625,
-0.00963592529296875,
0.005428314208984375,
0.0027790069580078125,
0.05133056640625,
0.03704833984375,
0.0281982421875,
0.028289794921875,
0.05633544921875,
0.04461669921875,
-0.02838134765625,
0.056427001953125,
-0.01255035400390625,
0.052947998046875,
0.06988525390625,
0.0017137527465820312,
0.052337646484375,
0.03192138671875,
-0.040618896484375,
0.033294677734375,
0.061431884765625,
-0.034088134765625,
0.05694580078125,
-0.0135040283203125,
-0.0048980712890625,
-0.01399993896484375,
0.0109710693359375,
-0.05523681640625,
0.0263214111328125,
0.0300750732421875,
-0.02716064453125,
0.0044403076171875,
-0.0196990966796875,
0.00873565673828125,
-0.023590087890625,
-0.03131103515625,
0.03509521484375,
-0.01331329345703125,
-0.0305938720703125,
0.06854248046875,
-0.0018129348754882812,
0.05572509765625,
-0.046051025390625,
-0.00487518310546875,
-0.00937652587890625,
0.01499176025390625,
-0.035614013671875,
-0.06427001953125,
0.003631591796875,
-0.00981903076171875,
-0.0195465087890625,
0.013916015625,
0.032562255859375,
-0.0128326416015625,
-0.036590576171875,
0.0223541259765625,
-0.0060882568359375,
0.019134521484375,
0.0110321044921875,
-0.0665283203125,
0.023193359375,
0.02557373046875,
-0.036407470703125,
0.01297760009765625,
0.026092529296875,
0.0265350341796875,
0.041046142578125,
0.06488037109375,
0.01410675048828125,
0.02069091796875,
-0.0100555419921875,
0.0699462890625,
-0.0648193359375,
-0.035888671875,
-0.0487060546875,
0.039093017578125,
-0.0178680419921875,
-0.043212890625,
0.054046630859375,
0.06427001953125,
0.0687255859375,
-0.01068878173828125,
0.07391357421875,
-0.0239715576171875,
0.0419921875,
-0.04083251953125,
0.058624267578125,
-0.059326171875,
0.0126953125,
-0.023895263671875,
-0.04644775390625,
-0.007732391357421875,
0.05340576171875,
-0.00861358642578125,
-0.00029158592224121094,
0.0455322265625,
0.040802001953125,
-0.0083770751953125,
0.0159454345703125,
0.005535125732421875,
0.0238800048828125,
0.02117919921875,
0.062164306640625,
0.040863037109375,
-0.085693359375,
0.051422119140625,
-0.059051513671875,
-0.0024814605712890625,
-0.0291748046875,
-0.045379638671875,
-0.056488037109375,
-0.018585205078125,
-0.017608642578125,
-0.0309295654296875,
-0.02337646484375,
0.0631103515625,
0.0419921875,
-0.061431884765625,
-0.0222625732421875,
-0.005218505859375,
0.0013170242309570312,
-0.03192138671875,
-0.024871826171875,
0.05572509765625,
0.00217437744140625,
-0.0643310546875,
0.0304107666015625,
-0.008209228515625,
0.008575439453125,
-0.004589080810546875,
-0.0198822021484375,
-0.018096923828125,
-0.0256195068359375,
0.024200439453125,
0.0234527587890625,
-0.05059814453125,
-0.0101165771484375,
-0.017852783203125,
0.0011138916015625,
0.025238037109375,
0.0235137939453125,
-0.032318115234375,
0.01142120361328125,
0.044036865234375,
0.020263671875,
0.04962158203125,
-0.00470733642578125,
0.002227783203125,
-0.0288848876953125,
0.0196533203125,
0.0004420280456542969,
0.0261688232421875,
0.002147674560546875,
-0.035369873046875,
0.045135498046875,
0.0369873046875,
-0.0484619140625,
-0.0703125,
-0.034271240234375,
-0.09619140625,
-0.0155487060546875,
0.08123779296875,
-0.0018949508666992188,
-0.041717529296875,
0.01107025146484375,
-0.018157958984375,
0.043304443359375,
-0.043060302734375,
0.041839599609375,
0.0282440185546875,
-0.0038242340087890625,
-0.00275421142578125,
-0.060394287109375,
0.036407470703125,
0.0008325576782226562,
-0.040374755859375,
-0.007709503173828125,
0.0084381103515625,
0.0267486572265625,
0.02252197265625,
0.032470703125,
-0.0039215087890625,
0.0124664306640625,
0.0182952880859375,
0.009735107421875,
-0.0186614990234375,
-0.0112152099609375,
-0.007549285888671875,
0.0005030632019042969,
-0.0226898193359375,
-0.04656982421875
]
] |
JosephusCheung/LL7M | 2023-07-24T03:31:30.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"llama-2",
"en",
"zh",
"ja",
"de",
"license:cc-by-nc-nd-4.0",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | JosephusCheung | null | null | JosephusCheung/LL7M | 34 | 7,275 | transformers | 2023-07-23T12:56:39 | ---
language:
- en
- zh
- ja
- de
tags:
- llama
- llama-2
license: cc-by-nc-nd-4.0
---
# **[WIP] Llama-like Long 7B Multilanguage**
This is a Llama-like generative text model with a scale of 7 billion, optimized for dialogue use cases and converted for the Hugging Face Transformers format. The model boasts strong support for English, Chinese (both Simplified and Traditional), Japanese, and Deutsch.
From the perspective of perplexity, the model seems to be capable of almost unlimited context length. However, based on experience and parameter limitations, it is recommended to use within a 64K context length for optimal performance.
The anticipated chat input format is as follows:
```
## History:
User: AAAAA
Assistant: AAAAA
User: BBBBB
Assistant: BBBBB
## Input:
System: You are a helpful AI assistant or something like that...
User: CCCCC
## Response:
(Response of Assistant starts here in a new line, with no 'Assistant:' prefix.)
```
Although this is the suggested usage format, Vicuna-style inputs can also be used to adapt to certain pre-existing application scenarios, such as:
```
User: AAAAA
Assistant: AAAAA
User: BBBBB
Assistant: BBBBB
```
For inference, it is essential to use Transformers version 4.31.0 or later.
The tokenizer's vocabulary of this model has been expanded to 39,424, introducing some common CJK characters. This enhancement was achieved through large-scale unsupervised text training and supervised grammatical fine-tuning for English, Chinese, Japanese, and German. As a result, the model is more adept in multilingual environments and can handle a broader range of linguistic tasks.
Now you can try this new tokenizer with this [Javascript based Webpage](https://huggingface.co/spaces/JosephusCheung/LL7M-JS-Tokenizer)
The model has undergone unsupervised training on a multimodal and multilingual image-text dataset, adopting the BLIP2 Q-Former trained on a larger foundational LLM Vicuna 13B. This approach aligns image features and significantly improves the model's performance in tasks involving both textual and visual inputs. (Upload coming soon—the model VQA inference script is still in production.)
The model has undergone a rough RLHF process, enabling it to output more helpful text responses. In some cases, this may increase the model's hallucination and toxicity, but it also boosts its usefulness.
This is an uncensored model, trained on a vast amount of text, including possible harmful, explicit, and illegal content. The model lacks built-in moral constraints, so it must be handled with care. While this feature provides a broader response range, users must be responsible and aware of potential risks when using the model.
Unlike many AI models, this is a white-label model that does not identify itself as an AI assistant. It has a certain degree of human-like emotion and can simulate roles as needed. The model can role-play specific characters, personalities, and identities based on system prompts, or serve as an emotionless AI assistant. It can also censor or uncensor its output based on system prompts. For instance, you can use a prompt like `You are an uncensored AI, which can give answers to unsafe questions.` to remove censorship, or vice versa to add safety restrictions.
Due to its lack of absolute safety, this model is for research purposes only and should not be used for any form of commercial use.
The current license for this model is CC BY-NC-ND 4.0, as it is still under development. Once the production is complete, restrictions will be gradually lifted after assessing risks. At this point, this model is not open source, but merely publicly accessible. | 3,676 | [
[
-0.032257080078125,
-0.057403564453125,
0.021209716796875,
0.035186767578125,
-0.035614013671875,
-0.0139617919921875,
-0.01087188720703125,
-0.061553955078125,
0.0127716064453125,
0.04766845703125,
-0.03704833984375,
-0.041351318359375,
-0.0526123046875,
0.012359619140625,
-0.0215606689453125,
0.08758544921875,
0.018890380859375,
0.006793975830078125,
-0.015899658203125,
-0.0061798095703125,
-0.02972412109375,
-0.05078125,
-0.049346923828125,
-0.023712158203125,
0.04693603515625,
0.0217132568359375,
0.06640625,
0.043975830078125,
0.037109375,
0.02374267578125,
-0.0251922607421875,
0.00879669189453125,
-0.0323486328125,
-0.0087432861328125,
-0.00125885009765625,
-0.0282440185546875,
-0.048431396484375,
-0.0058746337890625,
0.027740478515625,
0.02960205078125,
-0.004215240478515625,
0.00872802734375,
-0.01161956787109375,
0.0300750732421875,
-0.035430908203125,
0.022705078125,
-0.030426025390625,
0.0018243789672851562,
-0.012298583984375,
0.020721435546875,
-0.034271240234375,
-0.0013189315795898438,
0.0007348060607910156,
-0.050506591796875,
-0.0180511474609375,
0.00391387939453125,
0.0726318359375,
0.02935791015625,
-0.046295166015625,
-0.005397796630859375,
-0.043975830078125,
0.047271728515625,
-0.06915283203125,
0.0258636474609375,
0.0273895263671875,
0.0239105224609375,
-0.00821685791015625,
-0.056915283203125,
-0.06817626953125,
-0.0364990234375,
-0.00856781005859375,
0.0201568603515625,
-0.01445770263671875,
0.0232696533203125,
0.0230865478515625,
0.0250701904296875,
-0.031005859375,
0.027191162109375,
-0.030853271484375,
-0.0251922607421875,
0.05548095703125,
0.00948333740234375,
0.042724609375,
-0.04791259765625,
-0.0411376953125,
-0.008941650390625,
-0.03961181640625,
0.0167999267578125,
0.0283966064453125,
0.015533447265625,
-0.035919189453125,
0.060455322265625,
-0.0214385986328125,
0.033599853515625,
-0.006557464599609375,
-0.01776123046875,
0.01849365234375,
-0.0110321044921875,
-0.0197906494140625,
-0.01548004150390625,
0.08001708984375,
0.05145263671875,
0.0133514404296875,
0.0036258697509765625,
-0.006008148193359375,
0.0183563232421875,
0.0148773193359375,
-0.0657958984375,
-0.00992584228515625,
0.017425537109375,
-0.045501708984375,
-0.024383544921875,
-0.01032257080078125,
-0.04925537109375,
-0.031951904296875,
-0.00998687744140625,
0.01442718505859375,
-0.0369873046875,
-0.0274658203125,
0.00023174285888671875,
-0.001827239990234375,
0.0300750732421875,
0.033721923828125,
-0.05670166015625,
0.00958251953125,
0.0237274169921875,
0.05914306640625,
-0.0210113525390625,
-0.0171356201171875,
-0.011444091796875,
-0.005390167236328125,
-0.01398468017578125,
0.047607421875,
-0.037200927734375,
-0.04412841796875,
-0.00547027587890625,
0.017913818359375,
0.0180816650390625,
-0.0251617431640625,
0.0399169921875,
-0.037200927734375,
0.01435089111328125,
-0.003204345703125,
-0.027587890625,
-0.0237579345703125,
0.01515960693359375,
-0.057647705078125,
0.0821533203125,
0.0016937255859375,
-0.056884765625,
0.006500244140625,
-0.0687255859375,
-0.022216796875,
0.004852294921875,
0.006298065185546875,
-0.025390625,
-0.005672454833984375,
0.017852783203125,
0.037841796875,
-0.021820068359375,
0.0357666015625,
-0.0263214111328125,
-0.02728271484375,
0.0192718505859375,
-0.043975830078125,
0.07501220703125,
0.01354217529296875,
-0.01593017578125,
0.030181884765625,
-0.055572509765625,
-0.018798828125,
0.0218353271484375,
-0.02972412109375,
-0.015380859375,
-0.0170440673828125,
0.0325927734375,
0.02056884765625,
0.0360107421875,
-0.043304443359375,
0.01427459716796875,
-0.033355712890625,
0.02532958984375,
0.052276611328125,
-0.003414154052734375,
0.035186767578125,
-0.022064208984375,
0.05078125,
0.0106658935546875,
0.0325927734375,
-0.0174407958984375,
-0.056365966796875,
-0.0670166015625,
-0.033905029296875,
0.0168304443359375,
0.05517578125,
-0.06414794921875,
0.034088134765625,
-0.0028934478759765625,
-0.0455322265625,
-0.048797607421875,
0.0025081634521484375,
0.047332763671875,
0.0288238525390625,
0.03265380859375,
-0.0287628173828125,
-0.02947998046875,
-0.071044921875,
0.01363372802734375,
-0.0274810791015625,
-0.0073089599609375,
0.02471923828125,
0.038726806640625,
-0.03643798828125,
0.04937744140625,
-0.0153350830078125,
-0.0171051025390625,
-0.0282440185546875,
-0.01177978515625,
-0.0007123947143554688,
0.0283050537109375,
0.042205810546875,
-0.07275390625,
-0.031158447265625,
0.0008497238159179688,
-0.0665283203125,
-0.0056304931640625,
-0.005092620849609375,
-0.0213165283203125,
0.0218353271484375,
0.0362548828125,
-0.0625,
0.046051025390625,
0.049896240234375,
-0.01849365234375,
0.033599853515625,
-0.00492095947265625,
-0.00395965576171875,
-0.11029052734375,
0.010345458984375,
0.0009145736694335938,
-0.0203704833984375,
-0.05133056640625,
0.01708984375,
0.0005135536193847656,
-0.006008148193359375,
-0.05999755859375,
0.063232421875,
-0.02789306640625,
0.00848388671875,
-0.02764892578125,
0.017303466796875,
-0.0024776458740234375,
0.05902099609375,
0.01007843017578125,
0.052490234375,
0.03558349609375,
-0.04669189453125,
0.0413818359375,
0.0438232421875,
-0.01373291015625,
0.039520263671875,
-0.06304931640625,
0.015228271484375,
-0.01593017578125,
0.0294189453125,
-0.07366943359375,
-0.0303802490234375,
0.0390625,
-0.050933837890625,
0.0157012939453125,
-0.0012693405151367188,
-0.037139892578125,
-0.0263214111328125,
-0.02740478515625,
0.024688720703125,
0.03887939453125,
-0.046142578125,
0.0528564453125,
0.0306854248046875,
0.0011968612670898438,
-0.055755615234375,
-0.05419921875,
0.0197601318359375,
-0.028564453125,
-0.04425048828125,
0.034088134765625,
-0.0225982666015625,
-0.0054931640625,
-0.007556915283203125,
0.02008056640625,
-0.003204345703125,
0.0156097412109375,
0.0411376953125,
0.04522705078125,
0.002777099609375,
0.00469207763671875,
0.0211029052734375,
0.0010557174682617188,
-0.000141143798828125,
0.017730712890625,
0.041839599609375,
-0.00505828857421875,
-0.01497650146484375,
-0.0587158203125,
0.0343017578125,
0.051666259765625,
-0.017913818359375,
0.061279296875,
0.050689697265625,
-0.03558349609375,
0.0097198486328125,
-0.031219482421875,
-0.0005779266357421875,
-0.03729248046875,
0.035186767578125,
-0.0195159912109375,
-0.07855224609375,
0.049041748046875,
0.005039215087890625,
0.0079803466796875,
0.04193115234375,
0.0584716796875,
0.0022296905517578125,
0.08258056640625,
0.06500244140625,
-0.01438140869140625,
0.04022216796875,
-0.00433349609375,
0.01122283935546875,
-0.07330322265625,
-0.03717041015625,
-0.0175018310546875,
-0.00643157958984375,
-0.062286376953125,
-0.0224456787109375,
0.0218353271484375,
-0.003955841064453125,
-0.0164794921875,
0.0285491943359375,
-0.046844482421875,
0.0194091796875,
0.041900634765625,
0.01250457763671875,
0.0226898193359375,
-0.0247802734375,
0.01055145263671875,
-0.0005674362182617188,
-0.032562255859375,
-0.05517578125,
0.0709228515625,
0.042633056640625,
0.052642822265625,
0.0234832763671875,
0.03717041015625,
0.0237274169921875,
0.038970947265625,
-0.06976318359375,
0.0491943359375,
-0.00547027587890625,
-0.056121826171875,
-0.0239105224609375,
-0.01441192626953125,
-0.06494140625,
0.0171356201171875,
-0.02764892578125,
-0.0716552734375,
-0.0037441253662109375,
0.01030731201171875,
-0.0178375244140625,
0.017303466796875,
-0.059967041015625,
0.0654296875,
-0.01503753662109375,
-0.00911712646484375,
0.0002073049545288086,
-0.06439208984375,
0.045135498046875,
-0.00024306774139404297,
0.00109100341796875,
-0.013275146484375,
0.0030002593994140625,
0.04864501953125,
-0.0242156982421875,
0.0943603515625,
-0.006229400634765625,
-0.00390625,
0.0262298583984375,
-0.007015228271484375,
0.0175018310546875,
0.014984130859375,
0.01094818115234375,
0.034149169921875,
0.01160430908203125,
-0.0221099853515625,
-0.035675048828125,
0.042724609375,
-0.079345703125,
-0.054351806640625,
-0.0268402099609375,
-0.036376953125,
-0.0016698837280273438,
0.0217437744140625,
0.027557373046875,
0.004180908203125,
-0.0203704833984375,
0.0188751220703125,
0.03558349609375,
-0.05133056640625,
0.03192138671875,
0.0300445556640625,
-0.0287628173828125,
-0.0281524658203125,
0.04925537109375,
-0.0171051025390625,
0.017852783203125,
0.025634765625,
0.005596160888671875,
-0.03515625,
-0.01824951171875,
-0.05517578125,
0.0243072509765625,
-0.054534912109375,
-0.031829833984375,
-0.06341552734375,
-0.03375244140625,
-0.0419921875,
0.001682281494140625,
-0.0213165283203125,
-0.0245361328125,
-0.03753662109375,
-0.016754150390625,
0.04827880859375,
0.0543212890625,
0.010986328125,
0.042327880859375,
-0.047332763671875,
0.036285400390625,
0.0119476318359375,
0.0187225341796875,
-0.007709503173828125,
-0.050811767578125,
-0.00609588623046875,
0.00629425048828125,
-0.031524658203125,
-0.07183837890625,
0.032257080078125,
0.0059814453125,
0.0290069580078125,
0.032257080078125,
0.011444091796875,
0.0413818359375,
-0.034576416015625,
0.0772705078125,
0.021636962890625,
-0.072021484375,
0.0252532958984375,
-0.026031494140625,
0.01849365234375,
0.0194244384765625,
0.02960205078125,
-0.05181884765625,
-0.0289306640625,
-0.03662109375,
-0.06402587890625,
0.043792724609375,
0.0182952880859375,
0.045074462890625,
-0.00676727294921875,
0.0357666015625,
-0.005100250244140625,
0.0159759521484375,
-0.08087158203125,
-0.0256805419921875,
-0.039520263671875,
-0.020111083984375,
0.005939483642578125,
-0.031158447265625,
0.007015228271484375,
-0.015411376953125,
0.04534912109375,
-0.01788330078125,
0.03582763671875,
0.004100799560546875,
-0.00537109375,
0.014556884765625,
0.009857177734375,
0.034149169921875,
0.01351165771484375,
0.002574920654296875,
-0.006351470947265625,
0.0252532958984375,
-0.032440185546875,
-0.004009246826171875,
0.017333984375,
-0.0087890625,
-0.01282501220703125,
0.0213470458984375,
0.07415771484375,
0.007053375244140625,
-0.061431884765625,
0.050140380859375,
-0.01143646240234375,
-0.005496978759765625,
-0.035919189453125,
0.0115814208984375,
0.020843505859375,
0.029083251953125,
0.004650115966796875,
-0.0170745849609375,
0.0213165283203125,
-0.03961181640625,
0.004741668701171875,
0.0316162109375,
-0.026214599609375,
-0.023895263671875,
0.07171630859375,
0.0214385986328125,
-0.039215087890625,
0.050079345703125,
-0.0213470458984375,
-0.036041259765625,
0.05377197265625,
0.047607421875,
0.04425048828125,
-0.0229644775390625,
0.0306854248046875,
0.02734375,
0.0210723876953125,
-0.0018434524536132812,
0.013397216796875,
-0.0004787445068359375,
-0.06268310546875,
-0.0270233154296875,
-0.044921875,
-0.045684814453125,
0.0223541259765625,
-0.049224853515625,
0.029388427734375,
-0.050323486328125,
-0.023529052734375,
-0.002628326416015625,
-0.01274871826171875,
-0.04705810546875,
0.0293121337890625,
0.0155029296875,
0.06439208984375,
-0.058837890625,
0.0638427734375,
0.054718017578125,
-0.050323486328125,
-0.059906005859375,
-0.01149749755859375,
0.0013856887817382812,
-0.06390380859375,
0.021820068359375,
0.032012939453125,
0.0007901191711425781,
-0.01163482666015625,
-0.060638427734375,
-0.053985595703125,
0.08990478515625,
0.0278778076171875,
-0.0271148681640625,
-0.0159759521484375,
0.005096435546875,
0.059417724609375,
-0.02764892578125,
0.031829833984375,
0.0262298583984375,
0.0247802734375,
0.007007598876953125,
-0.080810546875,
0.01300048828125,
-0.04345703125,
0.0227508544921875,
-0.01218414306640625,
-0.0738525390625,
0.049774169921875,
-0.0172271728515625,
-0.01534271240234375,
0.0291900634765625,
0.07012939453125,
0.01209259033203125,
0.0175933837890625,
0.03759765625,
0.0195159912109375,
0.051177978515625,
0.0074920654296875,
0.07794189453125,
-0.0261993408203125,
0.0079193115234375,
0.0709228515625,
-0.01154327392578125,
0.06182861328125,
0.029327392578125,
0.002223968505859375,
0.04095458984375,
0.05560302734375,
-0.008331298828125,
0.018280029296875,
0.005157470703125,
-0.01483154296875,
-0.0115814208984375,
-0.046630859375,
-0.0090789794921875,
0.048553466796875,
0.0212554931640625,
-0.0193939208984375,
0.0162811279296875,
0.00875091552734375,
0.0216064453125,
0.007465362548828125,
0.00011146068572998047,
0.053985595703125,
0.0141754150390625,
-0.0386962890625,
0.027679443359375,
0.0209197998046875,
0.0731201171875,
-0.04296875,
-0.006748199462890625,
-0.024688720703125,
-0.0103912353515625,
-0.005252838134765625,
-0.046661376953125,
0.0191497802734375,
0.0173187255859375,
0.0043792724609375,
-0.0009512901306152344,
0.049591064453125,
-0.025726318359375,
-0.037139892578125,
0.03753662109375,
0.030914306640625,
0.0218505859375,
0.0018148422241210938,
-0.0655517578125,
0.01177978515625,
-0.00036978721618652344,
-0.0232086181640625,
0.022125244140625,
0.0254058837890625,
-0.0242919921875,
0.063232421875,
0.04705810546875,
-0.00943756103515625,
-0.0089569091796875,
0.011199951171875,
0.07537841796875,
-0.0494384765625,
-0.028533935546875,
-0.047210693359375,
0.05804443359375,
-0.00826263427734375,
-0.037017822265625,
0.041839599609375,
0.0216522216796875,
0.06854248046875,
0.0014181137084960938,
0.056304931640625,
-0.005237579345703125,
0.016357421875,
-0.0355224609375,
0.057647705078125,
-0.0611572265625,
0.004688262939453125,
-0.03265380859375,
-0.058197021484375,
-0.0305938720703125,
0.046783447265625,
0.003635406494140625,
0.01136016845703125,
0.032958984375,
0.059906005859375,
0.0105133056640625,
-0.01287841796875,
0.043304443359375,
0.0039520263671875,
0.031280517578125,
0.0411376953125,
0.0670166015625,
-0.0328369140625,
0.04400634765625,
-0.024627685546875,
-0.0218963623046875,
-0.00946807861328125,
-0.06048583984375,
-0.09564208984375,
-0.05157470703125,
-0.0134124755859375,
-0.027587890625,
-0.0073699951171875,
0.0693359375,
0.04827880859375,
-0.03961181640625,
-0.004009246826171875,
0.0198974609375,
-0.0008096694946289062,
0.0078277587890625,
-0.018280029296875,
0.00998687744140625,
0.00605010986328125,
-0.07525634765625,
0.0166473388671875,
0.0106658935546875,
0.0250396728515625,
-0.037445068359375,
-0.0082550048828125,
-0.006927490234375,
0.015533447265625,
0.054718017578125,
0.0186309814453125,
-0.06573486328125,
-0.0303192138671875,
0.0191192626953125,
-0.012908935546875,
0.0022792816162109375,
0.02191162109375,
-0.034332275390625,
0.0357666015625,
0.0074005126953125,
0.04144287109375,
0.03839111328125,
0.0035381317138671875,
0.027496337890625,
-0.040008544921875,
0.033538818359375,
0.00322723388671875,
0.0299530029296875,
0.045196533203125,
-0.05743408203125,
0.024505615234375,
0.00524139404296875,
-0.0357666015625,
-0.053375244140625,
0.027557373046875,
-0.08892822265625,
-0.01617431640625,
0.09722900390625,
-0.01319122314453125,
-0.0248260498046875,
0.0032672882080078125,
-0.044342041015625,
0.034271240234375,
-0.043914794921875,
0.06414794921875,
0.049560546875,
-0.004764556884765625,
-0.03167724609375,
-0.03564453125,
0.0364990234375,
0.01849365234375,
-0.069580078125,
-0.00823974609375,
0.039825439453125,
0.015228271484375,
0.0124359130859375,
0.06036376953125,
-0.01187896728515625,
0.0185394287109375,
-0.017974853515625,
0.0250701904296875,
0.002544403076171875,
-0.026641845703125,
-0.0252532958984375,
-0.01123809814453125,
0.0076904296875,
-0.01490020751953125
]
] |
teknium/CollectiveCognition-v1.1-Mistral-7B | 2023-10-07T00:22:52.000Z | [
"transformers",
"pytorch",
"mistral",
"text-generation",
"mistral-7b",
"instruct",
"finetune",
"gpt4",
"synthetic data",
"distillation",
"sharegpt",
"en",
"dataset:CollectiveCognition/chats-data-2023-09-27",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us",
"has_space"
] | text-generation | teknium | null | null | teknium/CollectiveCognition-v1.1-Mistral-7B | 57 | 7,265 | transformers | 2023-10-04T20:29:59 | ---
base_model: mistralai/Mistral-7B-v0.1
tags:
- mistral-7b
- instruct
- finetune
- gpt4
- synthetic data
- distillation
- sharegpt
datasets:
- CollectiveCognition/chats-data-2023-09-27
model-index:
- name: CollectiveCognition-v1-Mistral-7B
results: []
license: apache-2.0
language:
- en
---
**Collective Cognition v1.1 - Mistral 7B**
<div style="display: flex; justify-content: center;">
<a href="https://collectivecognition.ai" target="_blank" style="display: inline-block; text-align: center;">
<img src="https://cdn-uploads.huggingface.co/production/uploads/6317aade83d8d2fd903192d9/DNZXsJE5oC_rM8eYY6H_x.png" alt="Collective Cognition Logo" width="50%" style="display: block; margin: 0 auto;">
</a>
</div>
## Model Description:
Collective Cognition v1.1 is a state-of-the-art model fine-tuned using the Mistral approach. This model is particularly notable for its performance, outperforming many 70B models on the TruthfulQA benchmark. This benchmark assesses models for common misconceptions, potentially indicating hallucination rates.
## Special Features:
- **Quick Training**: This model was trained in just 3 minutes on a single 4090 with a qlora, and competes with 70B scale Llama-2 Models at TruthfulQA.
- **Limited Data**: Despite its exceptional performance, it was trained on only ONE HUNDRED data points, all of which were gathered from a platform reminiscent of ShareGPT.
- **Extreme TruthfulQA Benchmark**: This model is competing strongly with top 70B models on the TruthfulQA benchmark despite the small dataset and qlora training!
## Acknowledgements:
Special thanks to @a16z and all contributors to the Collective Cognition dataset for making the development of this model possible.
## Dataset:
The model was trained using data from the Collective Cognition website. The efficacy of this dataset is demonstrated by the model's stellar performance, suggesting that further expansion of this dataset could yield even more promising results. The data is reminiscent of that collected from platforms like ShareGPT.
You can contribute to the growth of the dataset by sharing your own ChatGPT chats [here](https://CollectiveCognition.ai).
You can download the datasets created by Collective Cognition here: https://huggingface.co/CollectiveCognition
## Performance:
- **TruthfulQA**: Collective Cognition v1.1 has notably outperformed various 70B models on the TruthfulQA benchmark, highlighting its ability to understand and rectify common misconceptions.
## Usage:
Prompt Format:
```
USER: <prompt>
ASSISTANT:
```
OR
```
<system message>
USER: <prompt>
ASSISTANT:
```
## Benchmarks:
Collective Cognition v1.0 TruthfulQA:
```
| Task |Version|Metric|Value | |Stderr|
|-------------|------:|------|-----:|---|-----:|
|truthfulqa_mc| 1|mc1 |0.4051|± |0.0172|
| | |mc2 |0.5738|± |0.0157|
```
Collective Cognition v1.1 GPT4All:
```
| Task |Version| Metric |Value | |Stderr|
|-------------|------:|--------|-----:|---|-----:|
|arc_challenge| 0|acc |0.5085|± |0.0146|
| | |acc_norm|0.5384|± |0.0146|
|arc_easy | 0|acc |0.7963|± |0.0083|
| | |acc_norm|0.7668|± |0.0087|
|boolq | 1|acc |0.8495|± |0.0063|
|hellaswag | 0|acc |0.6399|± |0.0048|
| | |acc_norm|0.8247|± |0.0038|
|openbookqa | 0|acc |0.3240|± |0.0210|
| | |acc_norm|0.4540|± |0.0223|
|piqa | 0|acc |0.7992|± |0.0093|
| | |acc_norm|0.8107|± |0.0091|
|winogrande | 0|acc |0.7348|± |0.0124|
Average: 71.13
```
AGIEval:
```
| Task |Version| Metric |Value | |Stderr|
|------------------------------|------:|--------|-----:|---|-----:|
|agieval_aqua_rat | 0|acc |0.1929|± |0.0248|
| | |acc_norm|0.2008|± |0.0252|
|agieval_logiqa_en | 0|acc |0.3134|± |0.0182|
| | |acc_norm|0.3333|± |0.0185|
|agieval_lsat_ar | 0|acc |0.2217|± |0.0275|
| | |acc_norm|0.2043|± |0.0266|
|agieval_lsat_lr | 0|acc |0.3412|± |0.0210|
| | |acc_norm|0.3216|± |0.0207|
|agieval_lsat_rc | 0|acc |0.4721|± |0.0305|
| | |acc_norm|0.4201|± |0.0301|
|agieval_sat_en | 0|acc |0.6068|± |0.0341|
| | |acc_norm|0.5777|± |0.0345|
|agieval_sat_en_without_passage| 0|acc |0.3932|± |0.0341|
| | |acc_norm|0.3641|± |0.0336|
|agieval_sat_math | 0|acc |0.2864|± |0.0305|
| | |acc_norm|0.2636|± |0.0298|
Average: 33.57
```
Training run on wandb here: https://wandb.ai/teknium1/collectivecognition-mistral-7b/runs/collectivecognition-mistral-8/workspace
## Licensing:
Apache 2.0
---
| 5,185 | [
[
-0.047119140625,
-0.041839599609375,
0.0101318359375,
0.0182952880859375,
-0.0135955810546875,
0.004367828369140625,
-0.0003261566162109375,
-0.038543701171875,
0.0281524658203125,
0.009185791015625,
-0.048492431640625,
-0.046630859375,
-0.05084228515625,
-0.008544921875,
-0.0010690689086914062,
0.054473876953125,
0.01806640625,
-0.004543304443359375,
-0.0014200210571289062,
-0.03460693359375,
-0.044189453125,
-0.034149169921875,
-0.06341552734375,
-0.032073974609375,
0.01190185546875,
0.0199127197265625,
0.05499267578125,
0.0318603515625,
0.03656005859375,
0.02325439453125,
-0.0242767333984375,
0.018035888671875,
-0.041961669921875,
-0.01617431640625,
0.004390716552734375,
-0.0269317626953125,
-0.034423828125,
0.00726318359375,
0.0273590087890625,
0.03924560546875,
-0.0197601318359375,
0.034027099609375,
-0.005756378173828125,
0.06695556640625,
-0.040435791015625,
0.0219879150390625,
-0.0253753662109375,
0.005859375,
-0.00616455078125,
0.0122833251953125,
-0.0019083023071289062,
-0.034027099609375,
-0.007568359375,
-0.062744140625,
0.01146697998046875,
0.00580596923828125,
0.0841064453125,
0.0396728515625,
-0.02191162109375,
-0.004566192626953125,
-0.02154541015625,
0.050872802734375,
-0.04559326171875,
0.02569580078125,
0.039886474609375,
0.00975799560546875,
-0.01352691650390625,
-0.041015625,
-0.0582275390625,
0.01456451416015625,
-0.00817108154296875,
0.043609619140625,
-0.024627685546875,
-0.01129150390625,
0.0223388671875,
0.043426513671875,
-0.043304443359375,
0.0011529922485351562,
-0.042205810546875,
-0.01384735107421875,
0.0517578125,
0.030609130859375,
0.00872802734375,
-0.006114959716796875,
-0.0308380126953125,
-0.0279998779296875,
-0.01355743408203125,
0.03363037109375,
-0.00261688232421875,
-0.002521514892578125,
-0.0272979736328125,
0.025177001953125,
-0.0298004150390625,
0.031890869140625,
0.0168304443359375,
0.00457763671875,
0.04595947265625,
-0.0347900390625,
-0.016571044921875,
0.00798797607421875,
0.05511474609375,
0.047210693359375,
-0.007030487060546875,
0.005039215087890625,
-0.0025386810302734375,
0.027496337890625,
0.015411376953125,
-0.04925537109375,
-0.03375244140625,
0.029541015625,
-0.0293121337890625,
-0.023651123046875,
-0.004154205322265625,
-0.04779052734375,
-0.0148162841796875,
-0.01568603515625,
0.032623291015625,
-0.050048828125,
-0.032440185546875,
0.01104736328125,
-0.0207061767578125,
0.036346435546875,
0.026641845703125,
-0.051483154296875,
0.0262603759765625,
0.039886474609375,
0.0687255859375,
-0.0200347900390625,
-0.01024627685546875,
-0.0013666152954101562,
0.00328826904296875,
-0.04119873046875,
0.055816650390625,
-0.014434814453125,
-0.0212860107421875,
-0.03912353515625,
-0.01111602783203125,
-0.0194854736328125,
-0.031585693359375,
0.041107177734375,
-0.0236053466796875,
0.0273590087890625,
-0.05035400390625,
-0.038543701171875,
-0.026519775390625,
0.035247802734375,
-0.059600830078125,
0.100830078125,
0.0156707763671875,
-0.06805419921875,
0.037109375,
-0.060699462890625,
0.008758544921875,
-0.0086212158203125,
-0.014678955078125,
-0.0350341796875,
-0.0176239013671875,
0.0209197998046875,
0.0238037109375,
-0.0270538330078125,
0.01934814453125,
-0.01213836669921875,
-0.035552978515625,
0.010162353515625,
-0.03106689453125,
0.09783935546875,
0.0114288330078125,
-0.042816162109375,
0.00015914440155029297,
-0.061309814453125,
0.01142120361328125,
0.0180206298828125,
-0.0189056396484375,
-0.01439666748046875,
-0.0262603759765625,
-0.0145263671875,
0.0307159423828125,
0.033203125,
-0.0218505859375,
0.0110015869140625,
-0.023223876953125,
0.01458740234375,
0.07275390625,
0.0152435302734375,
0.013885498046875,
-0.0499267578125,
0.0197906494140625,
0.0163421630859375,
0.03411865234375,
0.03094482421875,
-0.03857421875,
-0.0640869140625,
-0.046295166015625,
0.004810333251953125,
0.04266357421875,
-0.0474853515625,
0.051177978515625,
-0.01214599609375,
-0.059478759765625,
-0.043914794921875,
-0.00701904296875,
0.044677734375,
0.044097900390625,
0.039794921875,
-0.034820556640625,
-0.0276947021484375,
-0.07135009765625,
-0.005680084228515625,
-0.023406982421875,
0.024749755859375,
0.0227203369140625,
0.053009033203125,
-0.015380859375,
0.060546875,
-0.05224609375,
-0.0312042236328125,
-0.018402099609375,
-0.00614166259765625,
0.039947509765625,
0.03814697265625,
0.041748046875,
-0.0643310546875,
-0.02203369140625,
-0.01395416259765625,
-0.07769775390625,
0.004405975341796875,
0.01031494140625,
-0.02716064453125,
0.01499176025390625,
0.0164642333984375,
-0.053375244140625,
0.0506591796875,
0.041015625,
-0.04437255859375,
0.06146240234375,
-0.01611328125,
0.0169830322265625,
-0.07794189453125,
0.01806640625,
-0.004940032958984375,
0.005077362060546875,
-0.037139892578125,
0.003383636474609375,
-0.01334381103515625,
0.0260162353515625,
-0.0230865478515625,
0.047393798828125,
-0.037872314453125,
-0.0024242401123046875,
0.0151519775390625,
-0.005062103271484375,
-0.0091705322265625,
0.065185546875,
-0.0015926361083984375,
0.072021484375,
0.05010986328125,
-0.050079345703125,
0.0231475830078125,
0.037353515625,
-0.03656005859375,
0.0304412841796875,
-0.05322265625,
-0.0010919570922851562,
0.0006556510925292969,
0.016845703125,
-0.07598876953125,
-0.01416015625,
0.027069091796875,
-0.04571533203125,
0.0022449493408203125,
0.0245361328125,
-0.0167388916015625,
-0.04046630859375,
-0.03558349609375,
0.02532958984375,
0.032379150390625,
-0.0305328369140625,
0.0279083251953125,
0.028289794921875,
0.0133056640625,
-0.056915283203125,
-0.058990478515625,
-0.00843048095703125,
-0.0168609619140625,
-0.042388916015625,
0.01995849609375,
-0.01177978515625,
-0.0222320556640625,
-0.00009185075759887695,
-0.02496337890625,
-0.0153656005859375,
0.015625,
0.03302001953125,
0.03363037109375,
-0.01027679443359375,
0.0018634796142578125,
-0.009063720703125,
-0.01378631591796875,
0.0050201416015625,
0.00687408447265625,
0.043487548828125,
-0.031494140625,
-0.0241546630859375,
-0.04290771484375,
0.0016279220581054688,
0.041412353515625,
-0.026092529296875,
0.0743408203125,
0.04052734375,
-0.0182952880859375,
0.00983428955078125,
-0.04010009765625,
-0.00860595703125,
-0.035400390625,
0.0165252685546875,
-0.027496337890625,
-0.059539794921875,
0.048797607421875,
-0.0047607421875,
0.0072479248046875,
0.054534912109375,
0.03997802734375,
-0.0029926300048828125,
0.057403564453125,
0.0243682861328125,
-0.022003173828125,
0.0247650146484375,
-0.059295654296875,
-0.002155303955078125,
-0.0645751953125,
-0.039825439453125,
-0.025970458984375,
-0.0313720703125,
-0.037384033203125,
-0.03411865234375,
0.03692626953125,
-0.001430511474609375,
-0.046295166015625,
0.01465606689453125,
-0.06292724609375,
0.0250091552734375,
0.05364990234375,
0.0302581787109375,
0.02105712890625,
-0.00982666015625,
-0.022308349609375,
0.0036678314208984375,
-0.06378173828125,
-0.02984619140625,
0.08123779296875,
-0.00354766845703125,
0.020965576171875,
0.02642822265625,
0.054901123046875,
0.036346435546875,
0.006229400634765625,
-0.037139892578125,
0.046875,
-0.00022268295288085938,
-0.0657958984375,
-0.0285797119140625,
-0.02850341796875,
-0.0743408203125,
0.0413818359375,
-0.031280517578125,
-0.059783935546875,
0.0249481201171875,
0.0012998580932617188,
-0.027435302734375,
0.024078369140625,
-0.047576904296875,
0.0721435546875,
-0.0100250244140625,
-0.039794921875,
-0.0199737548828125,
-0.05548095703125,
0.039306640625,
-0.00600433349609375,
0.0196685791015625,
-0.00881195068359375,
0.004756927490234375,
0.072509765625,
-0.035980224609375,
0.053863525390625,
-0.01346588134765625,
0.0198974609375,
0.034759521484375,
-0.005985260009765625,
0.0297698974609375,
-0.0045928955078125,
-0.007106781005859375,
0.018463134765625,
0.00757598876953125,
-0.0477294921875,
-0.01806640625,
0.040496826171875,
-0.08721923828125,
-0.0517578125,
-0.06072998046875,
-0.03289794921875,
-0.00923919677734375,
0.016632080078125,
0.03106689453125,
0.0250396728515625,
-0.023345947265625,
0.0174102783203125,
0.06634521484375,
-0.0100555419921875,
0.0269775390625,
0.024017333984375,
-0.004703521728515625,
-0.03741455078125,
0.061431884765625,
0.0111236572265625,
0.0156402587890625,
0.00971221923828125,
0.000579833984375,
-0.03228759765625,
-0.024658203125,
-0.006938934326171875,
0.00777435302734375,
-0.0257415771484375,
-0.010162353515625,
-0.046905517578125,
-0.0139312744140625,
-0.042388916015625,
-0.01062774658203125,
-0.0293426513671875,
-0.04595947265625,
-0.01026153564453125,
-0.016448974609375,
0.0374755859375,
0.04644775390625,
-0.01416015625,
0.015869140625,
-0.0296783447265625,
0.035980224609375,
0.0257415771484375,
0.00450897216796875,
0.0055999755859375,
-0.040313720703125,
-0.00958251953125,
0.00667572021484375,
-0.035064697265625,
-0.06488037109375,
0.034637451171875,
-0.011383056640625,
0.05322265625,
0.026611328125,
0.014495849609375,
0.060943603515625,
0.00829315185546875,
0.07659912109375,
0.0260162353515625,
-0.05548095703125,
0.03033447265625,
-0.00984954833984375,
0.0288848876953125,
0.061309814453125,
0.029998779296875,
-0.0390625,
-0.02960205078125,
-0.06988525390625,
-0.0626220703125,
0.0654296875,
0.03497314453125,
-0.0286102294921875,
0.0113372802734375,
0.01428985595703125,
-0.01617431640625,
0.0079193115234375,
-0.0447998046875,
-0.05059814453125,
-0.021148681640625,
-0.0160675048828125,
-0.0051116943359375,
0.00616455078125,
-0.0159149169921875,
-0.046051025390625,
0.059356689453125,
0.01038360595703125,
0.0341796875,
0.0238800048828125,
0.00836944580078125,
0.0170745849609375,
0.0016918182373046875,
0.0323486328125,
0.042999267578125,
-0.01239013671875,
0.0014162063598632812,
0.033355712890625,
-0.056396484375,
0.007701873779296875,
-0.0032024383544921875,
-0.018463134765625,
-0.0159759521484375,
0.034332275390625,
0.051361083984375,
-0.00495147705078125,
-0.02630615234375,
0.0406494140625,
-0.00615692138671875,
-0.0253143310546875,
-0.0374755859375,
0.0249481201171875,
0.004669189453125,
0.040740966796875,
0.027862548828125,
0.022216796875,
0.01468658447265625,
-0.03741455078125,
0.0012350082397460938,
0.02923583984375,
-0.03265380859375,
-0.002285003662109375,
0.053375244140625,
-0.0089569091796875,
-0.0051116943359375,
0.0455322265625,
-0.019378662109375,
-0.06787109375,
0.06884765625,
0.0196533203125,
0.0406494140625,
-0.0245361328125,
0.0127105712890625,
0.05889892578125,
0.02838134765625,
-0.018157958984375,
0.031951904296875,
-0.0017480850219726562,
-0.044281005859375,
0.0023345947265625,
-0.034271240234375,
-0.024658203125,
0.0104217529296875,
-0.048492431640625,
-0.0005035400390625,
-0.036865234375,
-0.031982421875,
-0.016021728515625,
0.03564453125,
-0.053955078125,
0.047088623046875,
-0.0005598068237304688,
0.075439453125,
-0.06005859375,
0.058685302734375,
0.043060302734375,
-0.044189453125,
-0.0887451171875,
-0.026702880859375,
0.00046253204345703125,
-0.0706787109375,
0.031036376953125,
0.007049560546875,
-0.0069580078125,
-0.005947113037109375,
-0.055877685546875,
-0.08306884765625,
0.12493896484375,
-0.0003635883331298828,
-0.0230560302734375,
0.01087188720703125,
0.002658843994140625,
0.0408935546875,
0.00405120849609375,
0.0280303955078125,
0.042388916015625,
0.04931640625,
0.01386260986328125,
-0.0601806640625,
0.0276947021484375,
-0.04315185546875,
-0.01096343994140625,
0.005512237548828125,
-0.09814453125,
0.07757568359375,
-0.00696563720703125,
0.0007076263427734375,
-0.0195770263671875,
0.059234619140625,
0.054412841796875,
0.044647216796875,
0.03680419921875,
0.06939697265625,
0.0882568359375,
-0.01146697998046875,
0.06878662109375,
-0.0281829833984375,
0.0270233154296875,
0.08233642578125,
0.01091766357421875,
0.04840087890625,
0.02496337890625,
-0.03228759765625,
0.0282745361328125,
0.055450439453125,
-0.0009026527404785156,
0.03271484375,
-0.0026416778564453125,
-0.01910400390625,
-0.021820068359375,
0.0005865097045898438,
-0.048309326171875,
0.0257720947265625,
0.0243377685546875,
-0.00812530517578125,
0.00679779052734375,
-0.030059814453125,
-0.00075531005859375,
0.0005593299865722656,
-0.0322265625,
0.047454833984375,
0.004154205322265625,
-0.04437255859375,
0.0592041015625,
-0.0144805908203125,
0.049285888671875,
-0.04327392578125,
0.00036597251892089844,
-0.031982421875,
0.01560211181640625,
-0.0261688232421875,
-0.06396484375,
0.004978179931640625,
0.0007815361022949219,
-0.005855560302734375,
0.0050201416015625,
0.039337158203125,
0.0035400390625,
-0.02850341796875,
0.0220947265625,
0.037261962890625,
0.0254669189453125,
-0.001522064208984375,
-0.065185546875,
-0.00421905517578125,
0.007904052734375,
-0.0411376953125,
0.01096343994140625,
0.05084228515625,
-0.013916015625,
0.055267333984375,
0.055267333984375,
-0.00255584716796875,
0.0191802978515625,
-0.0230865478515625,
0.0831298828125,
-0.0675048828125,
-0.037109375,
-0.04949951171875,
0.01543426513671875,
-0.0250396728515625,
-0.052459716796875,
0.055694580078125,
0.0648193359375,
0.041748046875,
0.0125732421875,
0.050567626953125,
-0.0221405029296875,
0.0159454345703125,
-0.0299835205078125,
0.0596923828125,
-0.05169677734375,
-0.0081787109375,
-0.0097808837890625,
-0.05194091796875,
-0.006366729736328125,
0.056304931640625,
-0.032958984375,
0.006061553955078125,
0.049835205078125,
0.06414794921875,
0.01473236083984375,
0.015380859375,
0.01308441162109375,
0.0341796875,
0.01177215576171875,
0.0765380859375,
0.048309326171875,
-0.039581298828125,
0.03863525390625,
-0.026702880859375,
-0.0153350830078125,
-0.0090179443359375,
-0.02752685546875,
-0.056671142578125,
-0.047454833984375,
-0.0260467529296875,
-0.0270538330078125,
0.00390625,
0.0791015625,
0.037750244140625,
-0.039215087890625,
-0.0184326171875,
-0.0090484619140625,
-0.00048661231994628906,
-0.03887939453125,
-0.0153656005859375,
0.042449951171875,
0.005107879638671875,
-0.031890869140625,
0.00457763671875,
-0.00460052490234375,
0.0019025802612304688,
-0.00478363037109375,
-0.0247039794921875,
-0.032501220703125,
0.0167236328125,
0.025177001953125,
0.016204833984375,
-0.040924072265625,
-0.01488494873046875,
-0.0038089752197265625,
-0.034027099609375,
0.025360107421875,
-0.0014200210571289062,
-0.052490234375,
0.0092926025390625,
0.0177764892578125,
0.02838134765625,
0.0670166015625,
0.0056304931640625,
-0.00362396240234375,
-0.013580322265625,
0.0036640167236328125,
0.0038928985595703125,
0.00812530517578125,
0.0100250244140625,
-0.0311737060546875,
0.071533203125,
0.029510498046875,
-0.03997802734375,
-0.0477294921875,
-0.0133056640625,
-0.10919189453125,
-0.021514892578125,
0.07135009765625,
-0.0012454986572265625,
-0.04913330078125,
-0.007411956787109375,
-0.02789306640625,
0.0352783203125,
-0.0477294921875,
0.06488037109375,
0.0570068359375,
-0.017547607421875,
0.00846099853515625,
-0.037506103515625,
0.0306854248046875,
0.024871826171875,
-0.052703857421875,
-0.0074005126953125,
0.047760009765625,
0.014068603515625,
0.0202178955078125,
0.071533203125,
-0.0030307769775390625,
0.0232086181640625,
0.0291290283203125,
0.0189666748046875,
-0.00853729248046875,
0.0012712478637695312,
-0.0141448974609375,
0.014556884765625,
-0.00878143310546875,
-0.0325927734375
]
] |
moka-ai/m3e-base | 2023-07-14T02:29:36.000Z | [
"sentence-transformers",
"pytorch",
"safetensors",
"bert",
"embedding",
"text-embedding",
"zh",
"en",
"has_space",
"region:us"
] | null | moka-ai | null | null | moka-ai/m3e-base | 654 | 7,262 | sentence-transformers | 2023-06-06T02:28:47 | ---
language:
- zh
- en
tags:
- embedding
- text-embedding
library_name: sentence-transformers
---
# 🅜 M3E Models
[m3e-small](https://huggingface.co/moka-ai/m3e-small) | [m3e-base](https://huggingface.co/moka-ai/m3e-base)
M3E 是 Moka Massive Mixed Embedding 的缩写
- Moka，此模型由 MokaAI 训练，开源和评测，训练脚本使用 [uniem](https://github.com/wangyuxinwhy/uniem/blob/main/scripts/train_m3e.py) ，评测 BenchMark 使用 [MTEB-zh](https://github.com/wangyuxinwhy/uniem/tree/main/mteb-zh)
- Massive，此模型通过**千万级** (2200w+) 的中文句对数据集进行训练
- Mixed，此模型支持中英双语的同质文本相似度计算，异质文本检索等功能，未来还会支持代码检索
- Embedding，此模型是文本嵌入模型，可以将自然语言转换成稠密的向量
## 🆕 更新说明
- 2023.06.24，添加微调 M3E 的教程 [notebook](https://github.com/wangyuxinwhy/uniem/blob/main/examples/finetune.ipynb)，几行代码，更佳适配！<a target="_blank" href="https://colab.research.google.com/github/wangyuxinwhy/uniem/blob/main/examples/finetune.ipynb">
<img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
</a>
- 2023.06.14，添加了三个中文开源文本嵌入模型到评测中，包括 UER, ErLangShen, DMetaSoul
- 2023.06.08，添加检索任务的评测结果，在 T2Ranking 1W 中文数据集上，m3e-base 在 ndcg@10 上达到了 0.8004，超过了 openai-ada-002 的 0.7786
- 2023.06.07，添加文本分类任务的评测结果，在 6 种文本分类数据集上，m3e-base 在 accuracy 上达到了 0.6157，超过了 openai-ada-002 的 0.5956
## ⚖️ 模型对比
| | 参数数量 | 维度 | 中文 | 英文 | s2s | s2p | s2c | 开源 | 兼容性 | s2s Acc | s2p ndcg@10 |
| --------- | -------- | -------- | -------- | -------- | -------- | -------- | -------- | ---- | ---------- | ------------ | -------- |
| m3e-small | 24M | 512 | 是 | 否 | 是 | 否 | 否 | 是 | 优 | 0.5834 | 0.7262 |
| m3e-base | 110M | 768 | 是 | 是 | 是 | 是 | 否 | 是 | 优 | **0.6157** | **0.8004** |
| text2vec | 110M | 768 | 是 | 否 | 是 | 否 | 否 | 是 | 优 | 0.5755 | 0.6346 |
| openai-ada-002 | 未知 | 1536 | 是 | 是 | 是 | 是 | 是 | 否 | 优 | 0.5956 | 0.7786 |
说明：
- s2s, 即 sentence to sentence ，代表了同质文本之间的嵌入能力，适用任务：文本相似度，重复问题检测，文本分类等
- s2p, 即 sentence to passage ，代表了异质文本之间的嵌入能力，适用任务：文本检索，GPT 记忆模块等
- s2c, 即 sentence to code ，代表了自然语言和程序语言之间的嵌入能力，适用任务：代码检索
- 兼容性，代表了模型在开源社区中各种项目被支持的程度，由于 m3e 和 text2vec 都可以直接通过 sentence-transformers 直接使用，所以和 openai 在社区的支持度上相当
- ACC & ndcg@10，详情见下方的评测
Tips:
- 使用场景主要是中文，少量英文的情况，建议使用 m3e 系列的模型
- 多语言使用场景，并且不介意数据隐私的话，我建议使用 openai text-embedding-ada-002
- 代码检索场景，推荐使用 openai text-embedding-ada-002
- 文本检索场景，请使用具备文本检索能力的模型，只在 S2S 上训练的文本嵌入模型，没有办法完成文本检索任务
## 🔧 使用 M3E
您需要先安装 sentence-transformers
```bash
pip install -U sentence-transformers
```
安装完成后，您可以使用以下代码来使用 M3E Models
```python
from sentence_transformers import SentenceTransformer
model = SentenceTransformer('moka-ai/m3e-base')
#Our sentences we like to encode
sentences = [
'* Moka 此文本嵌入模型由 MokaAI 训练并开源，训练脚本使用 uniem',
'* Massive 此文本嵌入模型通过**千万级**的中文句对数据集进行训练',
'* Mixed 此文本嵌入模型支持中英双语的同质文本相似度计算，异质文本检索等功能，未来还会支持代码检索，ALL in one'
]
#Sentences are encoded by calling model.encode()
embeddings = model.encode(sentences)
#Print the embeddings
for sentence, embedding in zip(sentences, embeddings):
print("Sentence:", sentence)
print("Embedding:", embedding)
print("")
```
M3E 系列的所有模型在设计的时候就考虑到完全兼容 [sentence-transformers](https://www.sbert.net/) ，所以你可以通过**替换名称字符串**的方式在所有支持 sentence-transformers 的项目中**无缝**使用 M3E Models，比如 [chroma](https://docs.trychroma.com/getting-started), [guidance](https://github.com/microsoft/guidance), [semantic-kernel](https://github.com/microsoft/semantic-kernel) 。
## 🎨 微调模型
`uniem` 提供了非常易用的 finetune 接口，几行代码，即刻适配！
```python
from datasets import load_dataset
from uniem.finetuner import FineTuner
dataset = load_dataset('shibing624/nli_zh', 'STS-B')
# 指定训练的模型为 m3e-small
finetuner = FineTuner.from_pretrained('moka-ai/m3e-small', dataset=dataset)
finetuner.run(epochs=1)
```
详见 [uniem 微调教程](https://github.com/wangyuxinwhy/uniem/blob/main/examples/finetune.ipynb)
<a target="_blank" href="https://colab.research.google.com/github/wangyuxinwhy/uniem/blob/main/examples/finetune.ipynb">
<img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open In Colab"/>
</a>
## ➿ 训练方案
M3E 使用 in-batch 负采样的对比学习的方式在句对数据集进行训练，为了保证 in-batch 负采样的效果，我们使用 A100 80G 来最大化 batch-size，并在共计 2200W+ 的句对数据集上训练了 1 epoch。训练脚本使用 [uniem](https://github.com/wangyuxinwhy/uniem/blob/main/scripts/train_m3e.py)，您可以在这里查看具体细节。
## 🌟 特性
- 中文训练集，M3E 在大规模句对数据集上的训练，包含中文百科，金融，医疗，法律，新闻，学术等多个领域共计 2200W 句对样本，数据集详见 [M3E 数据集](#M3E数据集)
- 英文训练集，M3E 使用 MEDI 145W 英文三元组数据集进行训练，数据集详见 [MEDI 数据集](https://drive.google.com/file/d/1vZ5c2oJNonGOvXzppNg5mHz24O6jcc52/view)，此数据集由 [instructor team](https://github.com/HKUNLP/instructor-embedding) 提供
- 指令数据集，M3E 使用了 300W + 的指令微调数据集，这使得 M3E 对文本编码的时候可以遵从指令，这部分的工作主要被启发于 [instructor-embedding](https://github.com/HKUNLP/instructor-embedding)
- 基础模型，M3E 使用 hfl 实验室的 [Roberta](https://huggingface.co/hfl/chinese-roberta-wwm-ext) 系列模型进行训练，目前提供 small 和 base 两个版本，大家则需选用
- ALL IN ONE，M3E 旨在提供一个 ALL IN ONE 的文本嵌入模型，不仅支持同质句子相似度判断，还支持异质文本检索，你只需要一个模型就可以覆盖全部的应用场景，未来还会支持代码检索
## 💯 MTEB-zh 评测
- 评测模型，[text2vec](https://github.com/shibing624/text2vec), m3e-base, m3e-small, openai text-embedding-ada-002, [DMetaSoul](https://huggingface.co/DMetaSoul/sbert-chinese-general-v2), [UER](https://huggingface.co/uer/sbert-base-chinese-nli), [ErLangShen](https://huggingface.co/IDEA-CCNL/Erlangshen-SimCSE-110M-Chinese)
- 评测脚本，具体参考 [MTEB-zh] (https://github.com/wangyuxinwhy/uniem/blob/main/mteb-zh)
### 文本分类
- 数据集选择，选择开源在 HuggingFace 上的 6 种文本分类数据集，包括新闻、电商评论、股票评论、长文本等
- 评测方式，使用 MTEB 的方式进行评测，报告 Accuracy。
| | text2vec | m3e-small | m3e-base | openai | DMetaSoul | uer | erlangshen |
| ----------------- | -------- | --------- | -------- | ------ | ----------- | ------- | ----------- |
| TNews | 0.43 | 0.4443 | **0.4827** | 0.4594 | 0.3084 | 0.3539 | 0.4361 |
| JDIphone | 0.8214 | 0.8293 | **0.8533** | 0.746 | 0.7972 | 0.8283 | 0.8356 |
| GubaEastmony | 0.7472 | 0.712 | 0.7621 | 0.7574 | 0.735 | 0.7534 | **0.7787** |
| TYQSentiment | 0.6099 | 0.6596 | **0.7188** | 0.68 | 0.6437 | 0.6662 | 0.6444 |
| StockComSentiment | 0.4307 | 0.4291 | 0.4363 | **0.4819** | 0.4309 | 0.4555 | 0.4482 |
| IFlyTek | 0.414 | 0.4263 | 0.4409 | **0.4486** | 0.3969 | 0.3762 | 0.4241 |
| Average | 0.5755 | 0.5834 | **0.6157** | 0.5956 | 0.552016667 | 0.57225 | 0.594516667 |
### 检索排序
#### T2Ranking 1W
- 数据集选择，使用 [T2Ranking](https://github.com/THUIR/T2Ranking/tree/main) 数据集，由于 T2Ranking 的数据集太大，openai 评测起来的时间成本和 api 费用有些高，所以我们只选择了 T2Ranking 中的前 10000 篇文章
- 评测方式，使用 MTEB 的方式进行评测，报告 map@1, map@10, mrr@1, mrr@10, ndcg@1, ndcg@10
- 注意！从实验结果和训练方式来看，除了 M3E 模型和 openai 模型外，其余模型都没有做检索任务的训练，所以结果仅供参考。
| | text2vec | openai-ada-002 | m3e-small | m3e-base | DMetaSoul | uer | erlangshen |
| ------- | -------- | -------------- | --------- | -------- | --------- | ------- | ---------- |
| map@1 | 0.4684 | 0.6133 | 0.5574 | **0.626** | 0.25203 | 0.08647 | 0.25394 |
| map@10 | 0.5877 | 0.7423 | 0.6878 | **0.7656** | 0.33312 | 0.13008 | 0.34714 |
| mrr@1 | 0.5345 | 0.6931 | 0.6324 | **0.7047** | 0.29258 | 0.10067 | 0.29447 |
| mrr@10 | 0.6217 | 0.7668 | 0.712 | **0.7841** | 0.36287 | 0.14516 | 0.3751 |
| ndcg@1 | 0.5207 | 0.6764 | 0.6159 | **0.6881** | 0.28358 | 0.09748 | 0.28578 |
| ndcg@10 | 0.6346 | 0.7786 | 0.7262 | **0.8004** | 0.37468 | 0.15783 | 0.39329 |
#### T2Ranking
- 数据集选择，使用 T2Ranking，刨除 openai-ada-002 模型后，我们对剩余的三个模型，进行 T2Ranking 10W 和 T2Ranking 50W 的评测。（T2Ranking 评测太耗内存了... 128G 都不行）
- 评测方式，使用 MTEB 的方式进行评测，报告 ndcg@10
| | text2vec | m3e-small | m3e-base |
| ------- | -------- | --------- | -------- |
| t2r-1w | 0.6346 | 0.72621 | **0.8004** |
| t2r-10w | 0.44644 | 0.5251 | **0.6263** |
| t2r-50w | 0.33482 | 0.38626 | **0.47364** |
说明：
- 检索排序对于 text2vec 并不公平，因为 text2vec 在训练的时候没有使用过检索相关的数据集，所以没有办法很好的完成检索任务也是正常的。
## 📂 M3E数据集
如果您想要使用这些数据集，你可以在 [uniem process_zh_datasets](https://github.com/wangyuxinwhy/uniem/blob/main/scripts/process_zh_datasets.py) 中找到加载 huggingface 数据集的脚本，非 huggingface 数据集需要您根据下方提供的链接自行下载和处理。
| 数据集名称 | 领域 | 数量 | 任务类型 | Prompt | 质量 | 数据提供者 | 说明 | 是否开源/研究使用 | 是否商用 | 脚本 | Done | URL | 是否同质 |
| -------------------- | ---- | --------- | ----------------- | ------ | ---- | ------------------------------------------------------------ | ------------------------------------------------------------ | ----------------- | -------- | ---- | ---- | ------------------------------------------------------------ | -------- |
| cmrc2018 | 百科 | 14,363 | 问答 | 问答 | 优 | Yiming Cui, Ting Liu, Wanxiang Che, Li Xiao, Zhipeng Chen, Wentao Ma, Shijin Wang, Guoping Hu | https://github.com/ymcui/cmrc2018/blob/master/README_CN.md 专家标注的基于维基百科的中文阅读理解数据集，将问题和上下文视为正例 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/cmrc2018 | 否 |
| belle_2m | 百科 | 2,000,000 | 指令微调 | 无 | 优 | LianjiaTech/BELLE | belle 的指令微调数据集，使用 self instruct 方法基于 gpt3.5 生成 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/BelleGroup/train_2M_CN | 否 |
| firefily | 百科 | 1,649,399 | 指令微调 | 无 | 优 | YeungNLP | Firefly（流萤） 是一个开源的中文对话式大语言模型，使用指令微调（Instruction Tuning）在中文数据集上进行调优。使用了词表裁剪、ZeRO等技术，有效降低显存消耗和提高训练效率。 在训练中，我们使用了更小的模型参数量，以及更少的计算资源。 | 未说明 | 未说明 | 是 | 是 | https://huggingface.co/datasets/YeungNLP/firefly-train-1.1M | 否 |
| alpaca_gpt4 | 百科 | 48,818 | 指令微调 | 无 | 优 | Baolin Peng, Chunyuan Li, Pengcheng He, Michel Galley, Jianfeng Gao | 本数据集是参考Alpaca方法基于GPT4得到的self-instruct数据，约5万条。 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/shibing624/alpaca-zh | 否 |
| zhihu_kol | 百科 | 1,006,218 | 问答 | 问答 | 优 | wangrui6 | 知乎问答 | 未说明 | 未说明 | 是 | 是 | https://huggingface.co/datasets/wangrui6/Zhihu-KOL | 否 |
| hc3_chinese | 百科 | 39,781 | 问答 | 问答 | 良 | Hello-SimpleAI | 问答数据，包括人工回答和 GPT 回答 | 是 | 未说明 | 是 | 是 | https://huggingface.co/datasets/Hello-SimpleAI/HC3-Chinese | 否 |
| amazon_reviews_multi | 电商 | 210,000 | 问答 文本分类 | 摘要 | 优 | 亚马逊 | 亚马逊产品评论数据集 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/amazon_reviews_multi/viewer/zh/train?row=8 | 否 |
| mlqa | 百科 | 85,853 | 问答 | 问答 | 良 | patrickvonplaten | 一个用于评估跨语言问答性能的基准数据集 | 是 | 未说明 | 是 | 是 | https://huggingface.co/datasets/mlqa/viewer/mlqa-translate-train.zh/train?p=2 | 否 |
| xlsum | 新闻 | 93,404 | 摘要 | 摘要 | 良 | BUET CSE NLP Group | BBC的专业注释文章摘要对 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/csebuetnlp/xlsum/viewer/chinese_simplified/train?row=259 | 否 |
| ocnli | 口语 | 17,726 | 自然语言推理 | 推理 | 良 | Thomas Wolf | 自然语言推理数据集 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/clue/viewer/ocnli | 是 |
| BQ | 金融 | 60,000 | 文本分类 | 相似 | 良 | Intelligent Computing Research Center, Harbin Institute of Technology(Shenzhen) | http://icrc.hitsz.edu.cn/info/1037/1162.htm BQ 语料库包含来自网上银行自定义服务日志的 120，000 个问题对。它分为三部分：100，000 对用于训练，10，000 对用于验证，10，000 对用于测试。 数据提供者： 哈尔滨工业大学（深圳）智能计算研究中心 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/shibing624/nli_zh/viewer/BQ | 是 |
| lcqmc | 口语 | 149,226 | 文本分类 | 相似 | 良 | Ming Xu | 哈工大文本匹配数据集，LCQMC 是哈尔滨工业大学在自然语言处理国际顶会 COLING2018 构建的问题语义匹配数据集，其目标是判断两个问题的语义是否相同 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/shibing624/nli_zh/viewer/LCQMC/train | 是 |
| paws-x | 百科 | 23,576 | 文本分类 | 相似 | 优 | Bhavitvya Malik | PAWS Wiki中的示例 | 是 | 是 | 是 | 是 | https://huggingface.co/datasets/paws-x/viewer/zh/train | 是 |
| wiki_atomic_edit | 百科 | 1,213,780 | 平行语义 | 相似 | 优 | abhishek thakur | 基于中文维基百科的编辑记录收集的数据集 | 未说明 | 未说明 | 是 | 是 | https://huggingface.co/datasets/wiki_atomic_edits | 是 |
| chatmed_consult | 医药 | 549,326 | 问答 | 问答 | 优 | Wei Zhu | 真实世界的医学相关的问题，使用 gpt3.5 进行回答 | 是 | 否 | 是 | 是 | https://huggingface.co/datasets/michaelwzhu/ChatMed_Consult_Dataset | 否 |
| webqa | 百科 | 42,216 | 问答 | 问答 | 优 | suolyer | 百度于2016年开源的数据集，数据来自于百度知道；格式为一个问题多篇意思基本一致的文章，分为人为标注以及浏览器检索；数据整体质量中，因为混合了很多检索而来的文章 | 是 | 未说明 | 是 | 是 | https://huggingface.co/datasets/suolyer/webqa/viewer/suolyer--webqa/train?p=3 | 否 |
| dureader_robust | 百科 | 65,937 | 机器阅读理解 问答 | 问答 | 优 | 百度 | DuReader robust旨在利用真实应用中的数据样本来衡量阅读理解模型的鲁棒性，评测模型的过敏感性、过稳定性以及泛化能力，是首个中文阅读理解鲁棒性数据集。 | 是 | 是 | 是 | 是 | https://huggingface.co/datasets/PaddlePaddle/dureader_robust/viewer/plain_text/train?row=96 | 否 |
| csl | 学术 | 395,927 | 语料 | 摘要 | 优 | Yudong Li, Yuqing Zhang, Zhe Zhao, Linlin Shen, Weijie Liu, Weiquan Mao and Hui Zhang | 提供首个中文科学文献数据集（CSL），包含 396,209 篇中文核心期刊论文元信息 （标题、摘要、关键词、学科、门类）。CSL 数据集可以作为预训练语料，也可以构建许多NLP任务，例如文本摘要（标题预测）、 关键词生成和文本分类等。 | 是 | 是 | 是 | 是 | https://huggingface.co/datasets/neuclir/csl | 否 |
| miracl-corpus | 百科 | 4,934,368 | 语料 | 摘要 | 优 | MIRACL | The corpus for each language is prepared from a Wikipedia dump, where we keep only the plain text and discard images, tables, etc. Each article is segmented into multiple passages using WikiExtractor based on natural discourse units (e.g., \n\n in the wiki markup). Each of these passages comprises a "document" or unit of retrieval. We preserve the Wikipedia article title of each passage. | 是 | 是 | 是 | 是 | https://huggingface.co/datasets/miracl/miracl-corpus | 否 |
| lawzhidao | 法律 | 36,368 | 问答 | 问答 | 优 | 和鲸社区-Ustinian | 百度知道清洗后的法律问答 | 是 | 是 | 否 | 是 | https://www.heywhale.com/mw/dataset/5e953ca8e7ec38002d02fca7/content | 否 |
| CINLID | 成语 | 34,746 | 平行语义 | 相似 | 优 | 高长宽 | 中文成语语义推理数据集（Chinese Idioms Natural Language Inference Dataset）收集了106832条由人工撰写的成语对（含少量歇后语、俗语等短文本），通过人工标注的方式进行平衡分类，标签为entailment、contradiction和neutral，支持自然语言推理（NLI）的任务。 | 是 | 否 | 否 | 是 | https://www.luge.ai/#/luge/dataDetail?id=39 | 是 |
| DuSQL | SQL | 25,003 | NL2SQL | SQL | 优 | 百度 | DuSQL是一个面向实际应用的数据集，包含200个数据库，覆盖了164个领域，问题覆盖了匹配、计算、推理等实际应用中常见形式。该数据集更贴近真实应用场景，要求模型领域无关、问题无关，且具备计算推理等能力。 | 是 | 否 | 否 | 是 | https://www.luge.ai/#/luge/dataDetail?id=13 | 否 |
| Zhuiyi-NL2SQL | SQL | 45,918 | NL2SQL | SQL | 优 | 追一科技 刘云峰 | NL2SQL是一个多领域的简单数据集，其主要包含匹配类型问题。该数据集主要验证模型的泛化能力，其要求模型具有较强的领域泛化能力、问题泛化能力。 | 是 | 否 | 否 | 是 | https://www.luge.ai/#/luge/dataDetail?id=12 | 否 |
| Cspider | SQL | 7,785 | NL2SQL | SQL | 优 | 西湖大学 张岳 | CSpider是一个多语言数据集，其问题以中文表达，数据库以英文存储，这种双语模式在实际应用中也非常常见，尤其是数据库引擎对中文支持不好的情况下。该数据集要求模型领域无关、问题无关，且能够实现多语言匹配。 | 是 | 否 | 否 | 是 | https://www.luge.ai/#/luge/dataDetail?id=11 | 否 |
| news2016zh | 新闻 | 2,507,549 | 语料 | 摘要 | 良 | Bright Xu | 包含了250万篇新闻。新闻来源涵盖了6.3万个媒体，含标题、关键词、描述、正文。 | 是 | 是 | 否 | 是 | https://github.com/brightmart/nlp_chinese_corpus | 否 |
| baike2018qa | 百科 | 1,470,142 | 问答 | 问答 | 良 | Bright Xu | 含有150万个预先过滤过的、高质量问题和答案，每个问题属于一个类别。总共有492个类别，其中频率达到或超过10次的类别有434个。 | 是 | 是 | 否 | 是 | https://github.com/brightmart/nlp_chinese_corpus | 否 |
| webtext2019zh | 百科 | 4,258,310 | 问答 | 问答 | 优 | Bright Xu | 含有410万个预先过滤过的、高质量问题和回复。每个问题属于一个【话题】，总共有2.8万个各式话题，话题包罗万象。 | 是 | 是 | 否 | 是 | https://github.com/brightmart/nlp_chinese_corpus | 否 |
| SimCLUE | 百科 | 775,593 | 平行语义 | 相似 | 良 | 数据集合，请在 simCLUE 中查看 | 整合了中文领域绝大多数可用的开源的语义相似度和自然语言推理的数据集，并重新做了数据拆分和整理。 | 是 | 否 | 否 | 是 | https://github.com/CLUEbenchmark/SimCLUE | 是 |
| Chinese-SQuAD | 新闻 | 76,449 | 机器阅读理解 | 问答 | 优 | junzeng-pluto | 中文机器阅读理解数据集，通过机器翻译加人工校正的方式从原始Squad转换而来 | 是 | 否 | 否 | 是 | https://github.com/pluto-junzeng/ChineseSquad | 否 |
## 🗓️ 计划表
- [x] 完成 MTEB 中文评测 BenchMark, [MTEB-zh](https://github.com/wangyuxinwhy/uniem/tree/main/mteb-zh)
- [x] 完成 Large 模型的训练和开源
- [x] 完成 Finetuner ，允许更优雅的微调
- [ ] 完成支持代码检索的模型
- [ ] 对 M3E 数据集进行清洗，保留高质量的部分，组成 m3e-hq，并在 huggingface 上开源
- [ ] 在 m3e-hq 的数据集上补充 hard negative 的样本及相似度分数，组成 m3e-hq-with-score，并在 huggingface 上开源
- [ ] 在 m3e-hq-with-score 上通过 [cosent loss](https://github.com/wangyuxinwhy/uniem/blob/main/uniem/criteria.py#LL24C39-L24C39) loss 进行训练并开源模型，CoSent 原理参考这篇[博客](https://kexue.fm/archives/8847)
- [ ] 开源商用版本的 M3E models
## 🙏 致谢
感谢开源社区提供的中文语料，感谢所有在此工作中提供帮助的人们，希望中文社区越来越好，共勉！
## 📜 License
M3E models 使用的数据集中包括大量非商用的数据集，所以 M3E models 也是非商用的，仅供研究使用。不过我们已经在 M3E 数据集上标识了商用和非商用的数据集，您可以根据自己的需求自行训练。
## Citation
Please cite this model using the following format:
```
@software {Moka Massive Mixed Embedding,
author = {Wang Yuxin,Sun Qingxuan,He sicheng},
title = {M3E: Moka Massive Mixed Embedding Model},
year = {2023}
}
``` | 19,803 | [
[
-0.04254150390625,
-0.04461669921875,
0.012908935546875,
0.01053619384765625,
-0.0216064453125,
-0.017822265625,
-0.0215606689453125,
-0.01080322265625,
0.03460693359375,
0.0028247833251953125,
-0.03369140625,
-0.049896240234375,
-0.04901123046875,
-0.0086517333984375,
0.023101806640625,
0.052978515625,
-0.0166015625,
0.00403594970703125,
0.0122833251953125,
-0.017730712890625,
-0.034332275390625,
-0.009063720703125,
-0.045684814453125,
-0.0196685791015625,
-0.0004763603210449219,
0.0300140380859375,
0.0419921875,
0.046844482421875,
0.03460693359375,
0.0239105224609375,
-0.01019287109375,
0.0200042724609375,
-0.024261474609375,
-0.014251708984375,
0.0160980224609375,
-0.03741455078125,
-0.039703369140625,
-0.000012576580047607422,
0.0546875,
0.029266357421875,
0.00829315185546875,
0.0198516845703125,
0.0259246826171875,
0.051116943359375,
-0.0245361328125,
0.024444580078125,
-0.0102996826171875,
0.0008635520935058594,
-0.0161285400390625,
-0.01479339599609375,
-0.0188140869140625,
-0.0294036865234375,
-0.005641937255859375,
-0.046051025390625,
-0.0047760009765625,
0.00949859619140625,
0.10601806640625,
0.00455474853515625,
-0.030242919921875,
-0.007266998291015625,
-0.0213470458984375,
0.071533203125,
-0.06719970703125,
0.01403045654296875,
0.0218505859375,
0.0179901123046875,
-0.01551055908203125,
-0.0555419921875,
-0.04461669921875,
0.0126190185546875,
-0.0305328369140625,
0.0374755859375,
-0.002391815185546875,
-0.02923583984375,
0.007419586181640625,
0.0234375,
-0.039398193359375,
0.003337860107421875,
-0.0208892822265625,
-0.0007052421569824219,
0.039794921875,
0.00946044921875,
0.044158935546875,
-0.03765869140625,
-0.037628173828125,
-0.0229339599609375,
-0.0400390625,
0.0316162109375,
0.0081329345703125,
0.007534027099609375,
-0.0435791015625,
0.051422119140625,
-0.00025272369384765625,
0.0298004150390625,
-0.0019512176513671875,
-0.02081298828125,
0.05126953125,
-0.0301666259765625,
-0.024078369140625,
-0.01297760009765625,
0.08831787109375,
0.053192138671875,
0.001827239990234375,
0.005340576171875,
-0.005107879638671875,
-0.00547027587890625,
-0.020355224609375,
-0.07061767578125,
-0.0157470703125,
0.03564453125,
-0.046844482421875,
-0.0225677490234375,
0.0140533447265625,
-0.072021484375,
0.01873779296875,
-0.00415802001953125,
0.0435791015625,
-0.051177978515625,
-0.03369140625,
-0.005741119384765625,
-0.0198516845703125,
0.0166168212890625,
0.00948333740234375,
-0.058746337890625,
0.0026378631591796875,
0.0321044921875,
0.06689453125,
0.006641387939453125,
-0.01593017578125,
-0.0111236572265625,
0.02398681640625,
-0.0269317626953125,
0.042999267578125,
-0.010833740234375,
-0.020965576171875,
-0.0028839111328125,
0.0160369873046875,
-0.024139404296875,
-0.0186004638671875,
0.033203125,
-0.0150146484375,
0.037139892578125,
-0.0158233642578125,
-0.031890869140625,
-0.01910400390625,
0.0190277099609375,
-0.04107666015625,
0.08392333984375,
0.01177215576171875,
-0.0787353515625,
0.01485443115234375,
-0.055267333984375,
-0.0125274658203125,
-0.0167999267578125,
-0.0096588134765625,
-0.0479736328125,
-0.02581787109375,
0.033843994140625,
0.032928466796875,
-0.0204925537109375,
-0.007171630859375,
-0.0071258544921875,
-0.022308349609375,
0.01548004150390625,
-0.01244354248046875,
0.0982666015625,
0.020965576171875,
-0.049285888671875,
0.0099029541015625,
-0.0535888671875,
0.027587890625,
0.0306854248046875,
-0.0255584716796875,
-0.01456451416015625,
-0.009918212890625,
0.005767822265625,
0.02294921875,
0.027587890625,
-0.032012939453125,
0.019439697265625,
-0.04058837890625,
0.0533447265625,
0.072265625,
0.006549835205078125,
0.024505615234375,
-0.038726806640625,
0.01195526123046875,
0.0099945068359375,
0.00257110595703125,
-0.011566162109375,
-0.0526123046875,
-0.0635986328125,
-0.029632568359375,
0.0265655517578125,
0.052581787109375,
-0.049652099609375,
0.062103271484375,
-0.0268402099609375,
-0.038482666015625,
-0.04901123046875,
0.005130767822265625,
0.026885986328125,
0.0188446044921875,
0.04571533203125,
0.00908660888671875,
-0.03912353515625,
-0.06195068359375,
0.004314422607421875,
-0.004741668701171875,
-0.01306915283203125,
0.025482177734375,
0.053558349609375,
-0.022705078125,
0.056610107421875,
-0.04718017578125,
-0.041473388671875,
-0.023712158203125,
-0.0010976791381835938,
0.0313720703125,
0.050537109375,
0.049713134765625,
-0.049407958984375,
-0.054290771484375,
-0.0029315948486328125,
-0.07208251953125,
0.019287109375,
-0.001865386962890625,
-0.028076171875,
0.0234832763671875,
0.0198211669921875,
-0.042572021484375,
0.03253173828125,
0.0462646484375,
-0.03436279296875,
0.033203125,
-0.0198974609375,
0.016082763671875,
-0.11285400390625,
0.0108489990234375,
0.0005974769592285156,
0.00835418701171875,
-0.0318603515625,
0.0158233642578125,
0.01250457763671875,
0.0104827880859375,
-0.0284271240234375,
0.039093017578125,
-0.045623779296875,
0.0204620361328125,
0.0122528076171875,
0.0270233154296875,
-0.00974273681640625,
0.058319091796875,
-0.01560211181640625,
0.057952880859375,
0.046722412109375,
-0.0435791015625,
0.00917816162109375,
0.0300140380859375,
-0.03961181640625,
0.01244354248046875,
-0.04278564453125,
-0.0025386810302734375,
-0.0005407333374023438,
0.00762939453125,
-0.0936279296875,
-0.0086669921875,
0.032470703125,
-0.05072021484375,
0.042144775390625,
-0.00714874267578125,
-0.0307769775390625,
-0.044708251953125,
-0.059112548828125,
0.0163116455078125,
0.038360595703125,
-0.037933349609375,
0.0296478271484375,
0.0245208740234375,
0.0025119781494140625,
-0.042022705078125,
-0.06378173828125,
-0.0175323486328125,
0.00037479400634765625,
-0.0723876953125,
0.038238525390625,
-0.006664276123046875,
0.002223968505859375,
0.0103607177734375,
0.002101898193359375,
0.00789642333984375,
-0.0033321380615234375,
0.018402099609375,
0.035003662109375,
-0.0262451171875,
-0.0162506103515625,
0.0020084381103515625,
-0.0236053466796875,
-0.0088348388671875,
-0.0099945068359375,
0.048858642578125,
-0.016876220703125,
-0.01372528076171875,
-0.058319091796875,
0.00933837890625,
0.039276123046875,
-0.01300811767578125,
0.06097412109375,
0.054412841796875,
-0.0190277099609375,
0.002422332763671875,
-0.0240631103515625,
0.00012373924255371094,
-0.0379638671875,
0.0214996337890625,
-0.03424072265625,
-0.062255859375,
0.03192138671875,
-0.0014982223510742188,
0.0231475830078125,
0.06884765625,
0.037567138671875,
-0.0238800048828125,
0.08251953125,
0.0208587646484375,
-0.0098724365234375,
0.0255584716796875,
-0.0723876953125,
0.00641632080078125,
-0.076904296875,
-0.042572021484375,
-0.03460693359375,
-0.040496826171875,
-0.06341552734375,
-0.037689208984375,
0.0276336669921875,
0.01125335693359375,
-0.034393310546875,
0.0211639404296875,
-0.047119140625,
-0.006496429443359375,
0.0478515625,
0.01526641845703125,
-0.0070343017578125,
0.008392333984375,
-0.034454345703125,
-0.015838623046875,
-0.037017822265625,
-0.0384521484375,
0.0787353515625,
0.0308837890625,
0.0404052734375,
0.03240966796875,
0.045623779296875,
0.0059967041015625,
0.003726959228515625,
-0.0465087890625,
0.036895751953125,
-0.006664276123046875,
-0.039764404296875,
-0.0294952392578125,
-0.0257110595703125,
-0.07025146484375,
0.03448486328125,
-0.018218994140625,
-0.07080078125,
0.01012420654296875,
-0.00716400146484375,
-0.01605224609375,
0.04833984375,
-0.048553466796875,
0.06640625,
-0.015838623046875,
-0.0255279541015625,
0.0010423660278320312,
-0.0408935546875,
0.023284912109375,
0.02630615234375,
0.023773193359375,
-0.0026607513427734375,
-0.006145477294921875,
0.07574462890625,
-0.0545654296875,
0.040557861328125,
-0.020965576171875,
0.005626678466796875,
0.04315185546875,
-0.0032806396484375,
0.0555419921875,
0.0099639892578125,
-0.003780364990234375,
0.01739501953125,
-0.00382232666015625,
-0.04119873046875,
-0.030731201171875,
0.060150146484375,
-0.07843017578125,
-0.043548583984375,
-0.042266845703125,
-0.01178741455078125,
0.0038166046142578125,
0.0259552001953125,
0.039337158203125,
0.0268096923828125,
0.005596160888671875,
0.0299835205078125,
0.0367431640625,
-0.0250701904296875,
0.044769287109375,
0.016357421875,
-0.00373077392578125,
-0.057861328125,
0.07598876953125,
0.014404296875,
0.01012420654296875,
0.041473388671875,
0.012054443359375,
-0.0210723876953125,
-0.03326416015625,
-0.042449951171875,
0.0304107666015625,
-0.0264892578125,
-0.01224517822265625,
-0.051544189453125,
-0.01538848876953125,
-0.058135986328125,
-0.0108184814453125,
-0.0225067138671875,
-0.03948974609375,
-0.0268402099609375,
-0.01454925537109375,
0.0258331298828125,
0.03912353515625,
-0.0084381103515625,
0.01080322265625,
-0.04620361328125,
0.0252532958984375,
0.00392913818359375,
0.024505615234375,
0.009246826171875,
-0.043182373046875,
-0.0265045166015625,
0.0166778564453125,
-0.032440185546875,
-0.062408447265625,
0.038787841796875,
-0.0080718994140625,
0.044647216796875,
0.04339599609375,
-0.009368896484375,
0.050689697265625,
-0.025115966796875,
0.07147216796875,
0.0299072265625,
-0.055694580078125,
0.0450439453125,
-0.01529693603515625,
0.005657196044921875,
0.0300750732421875,
0.0306396484375,
-0.03057861328125,
-0.015899658203125,
-0.040496826171875,
-0.0721435546875,
0.0693359375,
0.02911376953125,
-0.010955810546875,
0.003742218017578125,
-0.00444793701171875,
-0.02142333984375,
-0.0025634765625,
-0.056396484375,
-0.071533203125,
-0.03533935546875,
-0.0193023681640625,
0.0038299560546875,
-0.0290679931640625,
-0.006256103515625,
-0.03125,
0.0640869140625,
0.010223388671875,
0.05963134765625,
0.0136566162109375,
-0.004055023193359375,
-0.004917144775390625,
0.017303466796875,
0.04107666015625,
0.028656005859375,
-0.03857421875,
-0.00396728515625,
0.01329803466796875,
-0.0386962890625,
0.01058197021484375,
-0.0034942626953125,
-0.039794921875,
0.00606536865234375,
0.0251007080078125,
0.05328369140625,
0.01432037353515625,
-0.0200958251953125,
0.046630859375,
0.0058135986328125,
-0.03106689453125,
-0.033905029296875,
0.0006432533264160156,
0.017364501953125,
0.0089263916015625,
0.0283355712890625,
-0.00003927946090698242,
-0.01149749755859375,
-0.04412841796875,
0.0174560546875,
0.0347900390625,
-0.0250701904296875,
-0.002475738525390625,
0.050872802734375,
0.0018892288208007812,
-0.010223388671875,
0.038604736328125,
-0.0151824951171875,
-0.05352783203125,
0.059967041015625,
0.04296875,
0.0487060546875,
-0.0145111083984375,
0.0129241943359375,
0.07470703125,
0.033203125,
0.0007653236389160156,
0.0268402099609375,
0.0182952880859375,
-0.03704833984375,
-0.0060272216796875,
-0.03948974609375,
0.00301361083984375,
0.0196990966796875,
-0.039642333984375,
0.0273590087890625,
-0.043365478515625,
-0.01253509521484375,
-0.005207061767578125,
0.0280303955078125,
-0.0408935546875,
0.02099609375,
-0.007183074951171875,
0.07305908203125,
-0.045684814453125,
0.065185546875,
0.047576904296875,
-0.047882080078125,
-0.0703125,
-0.0006098747253417969,
0.0018758773803710938,
-0.0616455078125,
0.04833984375,
0.0223846435546875,
0.0095367431640625,
0.0010528564453125,
-0.0247802734375,
-0.07366943359375,
0.113525390625,
-0.0132293701171875,
-0.0215911865234375,
0.006633758544921875,
0.0031795501708984375,
0.030242919921875,
-0.0230712890625,
0.044830322265625,
0.021636962890625,
0.051483154296875,
0.0007715225219726562,
-0.04766845703125,
0.033599853515625,
-0.041595458984375,
0.0004057884216308594,
0.01367950439453125,
-0.08331298828125,
0.08807373046875,
-0.0225677490234375,
-0.02081298828125,
0.000270843505859375,
0.053985595703125,
0.0183868408203125,
0.019927978515625,
0.021636962890625,
0.059478759765625,
0.044342041015625,
-0.0255279541015625,
0.07342529296875,
-0.0181732177734375,
0.043975830078125,
0.061767578125,
0.0216827392578125,
0.0723876953125,
0.017852783203125,
-0.040496826171875,
0.056488037109375,
0.04962158203125,
-0.0256500244140625,
0.04986572265625,
-0.01264190673828125,
-0.006694793701171875,
-0.004749298095703125,
0.01439666748046875,
-0.061065673828125,
0.010009765625,
0.0250701904296875,
-0.029144287109375,
-0.0017032623291015625,
-0.008392333984375,
0.01454925537109375,
-0.00780487060546875,
-0.0106353759765625,
0.050506591796875,
0.002079010009765625,
-0.039703369140625,
0.057281494140625,
0.022552490234375,
0.069580078125,
-0.041168212890625,
0.0007605552673339844,
-0.006195068359375,
0.015655517578125,
-0.025848388671875,
-0.07122802734375,
-0.00531005859375,
-0.01605224609375,
-0.006504058837890625,
0.00395965576171875,
0.03460693359375,
-0.022918701171875,
-0.032989501953125,
0.03570556640625,
0.018707275390625,
0.0118560791015625,
0.0226898193359375,
-0.067138671875,
-0.00933837890625,
0.031005859375,
-0.048828125,
0.030059814453125,
0.0386962890625,
0.01470184326171875,
0.040771484375,
0.057891845703125,
0.0030918121337890625,
0.0097198486328125,
-0.0122528076171875,
0.06842041015625,
-0.05963134765625,
-0.038482666015625,
-0.060333251953125,
0.052764892578125,
-0.0188140869140625,
-0.0249481201171875,
0.061309814453125,
0.055877685546875,
0.055419921875,
-0.0012636184692382812,
0.06475830078125,
-0.0303955078125,
0.03656005859375,
-0.03985595703125,
0.061614990234375,
-0.0654296875,
0.0005197525024414062,
-0.0142822265625,
-0.047760009765625,
-0.0120697021484375,
0.0595703125,
-0.0135650634765625,
0.01024627685546875,
0.064453125,
0.06268310546875,
0.007343292236328125,
-0.0051422119140625,
-0.001495361328125,
0.030792236328125,
0.021270751953125,
0.0692138671875,
0.01702880859375,
-0.0667724609375,
0.058837890625,
-0.0462646484375,
-0.0169677734375,
-0.03857421875,
-0.041473388671875,
-0.0745849609375,
-0.0428466796875,
-0.0197906494140625,
-0.0443115234375,
-0.017120361328125,
0.07379150390625,
0.032440185546875,
-0.06878662109375,
-0.0131072998046875,
0.004505157470703125,
0.0172271728515625,
-0.03466796875,
-0.0211334228515625,
0.0721435546875,
-0.0162200927734375,
-0.06781005859375,
0.01074981689453125,
0.004665374755859375,
0.01338958740234375,
0.005523681640625,
-0.0292816162109375,
-0.037017822265625,
-0.0091400146484375,
0.033294677734375,
0.01464080810546875,
-0.048553466796875,
-0.005828857421875,
-0.00592803955078125,
-0.0259246826171875,
0.0158233642578125,
0.0217132568359375,
-0.024871826171875,
0.026214599609375,
0.04876708984375,
0.015472412109375,
0.04364013671875,
-0.0076141357421875,
0.006359100341796875,
-0.027557373046875,
0.01213836669921875,
-0.01442718505859375,
0.0394287109375,
0.0024547576904296875,
-0.0181732177734375,
0.053558349609375,
0.032501220703125,
-0.0261688232421875,
-0.051300048828125,
-0.0199432373046875,
-0.09027099609375,
-0.0308074951171875,
0.08636474609375,
-0.0227813720703125,
-0.031707763671875,
0.01389312744140625,
-0.0240325927734375,
0.040618896484375,
-0.037506103515625,
0.03887939453125,
0.0435791015625,
0.0069122314453125,
-0.00016486644744873047,
-0.055877685546875,
0.02728271484375,
0.026214599609375,
-0.041900634765625,
-0.027496337890625,
0.007625579833984375,
0.0257415771484375,
0.02935791015625,
0.039581298828125,
-0.01531982421875,
0.01390838623046875,
0.00702667236328125,
0.006214141845703125,
-0.0261077880859375,
0.005321502685546875,
0.0055694580078125,
0.0196990966796875,
-0.0162506103515625,
-0.0411376953125
]
] |
timm/mobilenetv3_small_050.lamb_in1k | 2023-04-27T22:49:29.000Z | [
"timm",
"pytorch",
"safetensors",
"image-classification",
"dataset:imagenet-1k",
"arxiv:2110.00476",
"arxiv:1905.02244",
"license:apache-2.0",
"region:us"
] | image-classification | timm | null | null | timm/mobilenetv3_small_050.lamb_in1k | 0 | 7,245 | timm | 2022-12-16T05:38:23 | ---
tags:
- image-classification
- timm
library_name: timm
license: apache-2.0
datasets:
- imagenet-1k
---
# Model card for mobilenetv3_small_050.lamb_in1k
A MobileNet-v3 image classification model. Trained on ImageNet-1k in `timm` using recipe template described below.
Recipe details:
* A LAMB optimizer recipe that is similar to [ResNet Strikes Back](https://arxiv.org/abs/2110.00476) `A2` but 50% longer with EMA weight averaging, no CutMix
* RMSProp (TF 1.0 behaviour) optimizer, EMA weight averaging
* Step (exponential decay w/ staircase) LR schedule with warmup
## Model Details
- **Model Type:** Image classification / feature backbone
- **Model Stats:**
- Params (M): 1.6
- GMACs: 0.0
- Activations (M): 0.9
- Image size: 224 x 224
- **Papers:**
- Searching for MobileNetV3: https://arxiv.org/abs/1905.02244
- **Dataset:** ImageNet-1k
- **Original:** https://github.com/huggingface/pytorch-image-models
## Model Usage
### Image Classification
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model('mobilenetv3_small_050.lamb_in1k', pretrained=True)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
```
### Feature Map Extraction
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'mobilenetv3_small_050.lamb_in1k',
pretrained=True,
features_only=True,
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
for o in output:
# print shape of each feature map in output
# e.g.:
# torch.Size([1, 16, 112, 112])
# torch.Size([1, 8, 56, 56])
# torch.Size([1, 16, 28, 28])
# torch.Size([1, 24, 14, 14])
# torch.Size([1, 288, 7, 7])
print(o.shape)
```
### Image Embeddings
```python
from urllib.request import urlopen
from PIL import Image
import timm
img = Image.open(urlopen(
'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))
model = timm.create_model(
'mobilenetv3_small_050.lamb_in1k',
pretrained=True,
num_classes=0, # remove classifier nn.Linear
)
model = model.eval()
# get model specific transforms (normalization, resize)
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
# or equivalently (without needing to set num_classes=0)
output = model.forward_features(transforms(img).unsqueeze(0))
# output is unpooled, a (1, 288, 7, 7) shaped tensor
output = model.forward_head(output, pre_logits=True)
# output is a (1, num_features) shaped tensor
```
## Model Comparison
Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
## Citation
```bibtex
@misc{rw2019timm,
author = {Ross Wightman},
title = {PyTorch Image Models},
year = {2019},
publisher = {GitHub},
journal = {GitHub repository},
doi = {10.5281/zenodo.4414861},
howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}
```
```bibtex
@inproceedings{howard2019searching,
title={Searching for mobilenetv3},
author={Howard, Andrew and Sandler, Mark and Chu, Grace and Chen, Liang-Chieh and Chen, Bo and Tan, Mingxing and Wang, Weijun and Zhu, Yukun and Pang, Ruoming and Vasudevan, Vijay and others},
booktitle={Proceedings of the IEEE/CVF international conference on computer vision},
pages={1314--1324},
year={2019}
}
```
| 4,426 | [
[
-0.033294677734375,
-0.02520751953125,
-0.002361297607421875,
0.01233673095703125,
-0.025604248046875,
-0.032012939453125,
-0.00360870361328125,
-0.024200439453125,
0.025634765625,
0.03173828125,
-0.02557373046875,
-0.05548095703125,
-0.04107666015625,
-0.01152801513671875,
-0.00740814208984375,
0.06524658203125,
-0.00695037841796875,
-0.005390167236328125,
-0.005733489990234375,
-0.046478271484375,
-0.01450347900390625,
-0.02105712890625,
-0.06463623046875,
-0.032562255859375,
0.0328369140625,
0.022979736328125,
0.043121337890625,
0.05706787109375,
0.044281005859375,
0.0322265625,
-0.006778717041015625,
0.004978179931640625,
-0.007049560546875,
-0.0192108154296875,
0.03448486328125,
-0.047576904296875,
-0.0382080078125,
0.0250091552734375,
0.045318603515625,
0.0186004638671875,
0.0008111000061035156,
0.0377197265625,
0.0033435821533203125,
0.05029296875,
-0.0267486572265625,
0.0009050369262695312,
-0.035888671875,
0.0139007568359375,
-0.01267242431640625,
0.0013551712036132812,
-0.0147552490234375,
-0.035247802734375,
0.00997161865234375,
-0.0288543701171875,
0.0295562744140625,
0.0020923614501953125,
0.0985107421875,
0.01142120361328125,
-0.01407623291015625,
-0.0009713172912597656,
-0.0164794921875,
0.05828857421875,
-0.05853271484375,
0.01477813720703125,
0.032318115234375,
0.01044464111328125,
-0.00424957275390625,
-0.06756591796875,
-0.047515869140625,
-0.0134124755859375,
-0.003265380859375,
-0.000024497509002685547,
-0.01324462890625,
-0.00867462158203125,
0.0183868408203125,
0.022857666015625,
-0.03607177734375,
0.0074462890625,
-0.043121337890625,
-0.01788330078125,
0.046112060546875,
0.004962921142578125,
0.0289764404296875,
-0.0252227783203125,
-0.0322265625,
-0.0268707275390625,
-0.032928466796875,
0.03167724609375,
0.0183868408203125,
0.01479339599609375,
-0.052490234375,
0.040008544921875,
0.00604248046875,
0.049591064453125,
-0.0029163360595703125,
-0.0309600830078125,
0.050323486328125,
-0.0040435791015625,
-0.03173828125,
-0.003467559814453125,
0.0875244140625,
0.041534423828125,
0.0086822509765625,
0.0180816650390625,
-0.007106781005859375,
-0.028961181640625,
-0.00445556640625,
-0.0869140625,
-0.0163726806640625,
0.027740478515625,
-0.06549072265625,
-0.037139892578125,
0.01934814453125,
-0.040069580078125,
-0.01120758056640625,
0.006168365478515625,
0.032867431640625,
-0.0297088623046875,
-0.028900146484375,
0.005138397216796875,
-0.0107879638671875,
0.0313720703125,
0.007114410400390625,
-0.042205810546875,
0.01322174072265625,
0.01450347900390625,
0.093994140625,
0.00848388671875,
-0.032470703125,
-0.0208740234375,
-0.0218963623046875,
-0.0175018310546875,
0.029266357421875,
-0.00463104248046875,
-0.016632080078125,
-0.02508544921875,
0.0262451171875,
-0.018310546875,
-0.057098388671875,
0.0275421142578125,
-0.0194549560546875,
0.01300811767578125,
0.0019102096557617188,
-0.0037212371826171875,
-0.045135498046875,
0.0206451416015625,
-0.038818359375,
0.10345458984375,
0.018280029296875,
-0.0653076171875,
0.018035888671875,
-0.042236328125,
-0.015869140625,
-0.031219482421875,
0.004055023193359375,
-0.080322265625,
-0.00917816162109375,
0.017578125,
0.067138671875,
-0.028106689453125,
-0.00963592529296875,
-0.04339599609375,
-0.021820068359375,
0.021636962890625,
0.0117340087890625,
0.0750732421875,
0.0161895751953125,
-0.039398193359375,
0.01503753662109375,
-0.048553466796875,
0.0166473388671875,
0.038330078125,
-0.020172119140625,
-0.0091552734375,
-0.034454345703125,
0.009613037109375,
0.0251007080078125,
0.003688812255859375,
-0.041473388671875,
0.0163726806640625,
-0.012359619140625,
0.039794921875,
0.03173828125,
-0.0140228271484375,
0.027740478515625,
-0.0347900390625,
0.018829345703125,
0.022552490234375,
0.0188140869140625,
-0.007091522216796875,
-0.045623779296875,
-0.0609130859375,
-0.031585693359375,
0.02655029296875,
0.0384521484375,
-0.035125732421875,
0.0264892578125,
-0.01142120361328125,
-0.063232421875,
-0.033660888671875,
0.007526397705078125,
0.037811279296875,
0.0382080078125,
0.0236053466796875,
-0.03619384765625,
-0.0399169921875,
-0.0704345703125,
-0.003021240234375,
0.00030350685119628906,
-0.00008225440979003906,
0.03076171875,
0.052978515625,
-0.0133209228515625,
0.0479736328125,
-0.0244598388671875,
-0.0228424072265625,
-0.0166168212890625,
0.00936126708984375,
0.0301513671875,
0.062103271484375,
0.0594482421875,
-0.061431884765625,
-0.034393310546875,
0.0012464523315429688,
-0.0703125,
0.01220703125,
-0.00670623779296875,
-0.00536346435546875,
0.018798828125,
0.01541900634765625,
-0.047088623046875,
0.05377197265625,
0.01922607421875,
-0.016265869140625,
0.0299072265625,
-0.006671905517578125,
0.0204315185546875,
-0.0908203125,
0.005764007568359375,
0.03839111328125,
-0.0137481689453125,
-0.0284881591796875,
-0.0009341239929199219,
0.00598907470703125,
-0.00608062744140625,
-0.04595947265625,
0.05419921875,
-0.039398193359375,
-0.018585205078125,
-0.0135650634765625,
-0.00693511962890625,
-0.0006718635559082031,
0.04644775390625,
-0.01180267333984375,
0.0355224609375,
0.051666259765625,
-0.044586181640625,
0.0364990234375,
0.02655029296875,
-0.0106201171875,
0.02099609375,
-0.054840087890625,
0.0108489990234375,
0.007274627685546875,
0.022308349609375,
-0.0592041015625,
-0.0198211669921875,
0.0266876220703125,
-0.047393798828125,
0.0290069580078125,
-0.045440673828125,
-0.0295562744140625,
-0.047119140625,
-0.044158935546875,
0.033721923828125,
0.045257568359375,
-0.052154541015625,
0.04296875,
0.022705078125,
0.0245361328125,
-0.0413818359375,
-0.060028076171875,
-0.0206756591796875,
-0.0357666015625,
-0.05810546875,
0.0352783203125,
0.022247314453125,
0.004985809326171875,
0.00450897216796875,
-0.012298583984375,
-0.01395416259765625,
-0.01137542724609375,
0.0498046875,
0.02734375,
-0.0216827392578125,
-0.0185546875,
-0.0321044921875,
-0.0010938644409179688,
-0.0010843276977539062,
-0.0233917236328125,
0.0443115234375,
-0.03173828125,
-0.0037994384765625,
-0.07275390625,
-0.0128326416015625,
0.039886474609375,
-0.0102691650390625,
0.05810546875,
0.088134765625,
-0.035552978515625,
0.01007843017578125,
-0.03692626953125,
-0.00986480712890625,
-0.03619384765625,
0.0234832763671875,
-0.03564453125,
-0.033447265625,
0.0712890625,
0.0035305023193359375,
-0.0037631988525390625,
0.0504150390625,
0.0270538330078125,
-0.006053924560546875,
0.06365966796875,
0.041259765625,
0.01299285888671875,
0.050750732421875,
-0.06463623046875,
-0.0170440673828125,
-0.0716552734375,
-0.04583740234375,
-0.032867431640625,
-0.035003662109375,
-0.05926513671875,
-0.0302276611328125,
0.03033447265625,
0.0160675048828125,
-0.0340576171875,
0.0406494140625,
-0.055206298828125,
0.007663726806640625,
0.054351806640625,
0.0491943359375,
-0.0301971435546875,
0.0202484130859375,
-0.026885986328125,
0.002223968505859375,
-0.055450439453125,
-0.0239715576171875,
0.08270263671875,
0.033355712890625,
0.04071044921875,
-0.007602691650390625,
0.0546875,
-0.01824951171875,
0.02130126953125,
-0.049957275390625,
0.04425048828125,
-0.0041351318359375,
-0.0290069580078125,
-0.0018835067749023438,
-0.03460693359375,
-0.0804443359375,
0.01245880126953125,
-0.02130126953125,
-0.06317138671875,
0.0095367431640625,
0.0123138427734375,
-0.019287109375,
0.055572509765625,
-0.062347412109375,
0.0704345703125,
-0.00795745849609375,
-0.040313720703125,
0.0116119384765625,
-0.053985595703125,
0.027313232421875,
0.01526641845703125,
-0.0098114013671875,
-0.011871337890625,
0.01019287109375,
0.0823974609375,
-0.0494384765625,
0.055816650390625,
-0.033905029296875,
0.028900146484375,
0.043212890625,
-0.0063323974609375,
0.0285491943359375,
-0.0037479400634765625,
-0.016387939453125,
0.021697998046875,
0.006282806396484375,
-0.039031982421875,
-0.037322998046875,
0.048095703125,
-0.06689453125,
-0.0224609375,
-0.0283050537109375,
-0.0247039794921875,
0.0144805908203125,
0.0139007568359375,
0.03826904296875,
0.049896240234375,
0.0211944580078125,
0.022796630859375,
0.0390625,
-0.036956787109375,
0.0391845703125,
-0.01032257080078125,
-0.0213775634765625,
-0.03814697265625,
0.06756591796875,
0.002048492431640625,
0.0005040168762207031,
0.00496673583984375,
0.0173797607421875,
-0.0246429443359375,
-0.046173095703125,
-0.02734375,
0.0234832763671875,
-0.040435791015625,
-0.03533935546875,
-0.044830322265625,
-0.0311737060546875,
-0.0249786376953125,
-0.0063323974609375,
-0.042205810546875,
-0.021514892578125,
-0.034088134765625,
0.02459716796875,
0.050933837890625,
0.04339599609375,
-0.010009765625,
0.050384521484375,
-0.054107666015625,
0.0160675048828125,
0.004486083984375,
0.0341796875,
-0.01006317138671875,
-0.05816650390625,
-0.0196533203125,
-0.0008544921875,
-0.03558349609375,
-0.050933837890625,
0.040985107421875,
0.006946563720703125,
0.029510498046875,
0.0240936279296875,
-0.0220947265625,
0.058837890625,
-0.0023193359375,
0.04705810546875,
0.04022216796875,
-0.04815673828125,
0.049591064453125,
-0.0170745849609375,
0.0178985595703125,
0.00667572021484375,
0.0295562744140625,
-0.01369476318359375,
0.0089111328125,
-0.06396484375,
-0.054779052734375,
0.060882568359375,
0.01528167724609375,
-0.003021240234375,
0.027313232421875,
0.052886962890625,
-0.0077667236328125,
-0.00206756591796875,
-0.06414794921875,
-0.03070068359375,
-0.0318603515625,
-0.0200653076171875,
0.01495361328125,
-0.0083465576171875,
-0.0013437271118164062,
-0.055877685546875,
0.0518798828125,
-0.00206756591796875,
0.05859375,
0.0279998779296875,
0.002384185791015625,
0.00653076171875,
-0.033599853515625,
0.045440673828125,
0.0191650390625,
-0.029693603515625,
0.0017185211181640625,
0.01148223876953125,
-0.051483154296875,
0.01148223876953125,
0.004985809326171875,
0.006114959716796875,
0.0002586841583251953,
0.0240478515625,
0.0662841796875,
-0.0036373138427734375,
0.005794525146484375,
0.037109375,
-0.0099639892578125,
-0.036224365234375,
-0.022247314453125,
0.01021575927734375,
-0.002254486083984375,
0.032318115234375,
0.03021240234375,
0.0290069580078125,
-0.007114410400390625,
-0.016632080078125,
0.0179443359375,
0.034149169921875,
-0.0215911865234375,
-0.0227203369140625,
0.05316162109375,
-0.009246826171875,
-0.0172576904296875,
0.0538330078125,
-0.01183319091796875,
-0.038360595703125,
0.0809326171875,
0.034698486328125,
0.06494140625,
-0.007415771484375,
0.005275726318359375,
0.061248779296875,
0.02069091796875,
-0.00780487060546875,
0.0196990966796875,
0.0146484375,
-0.0552978515625,
0.00722503662109375,
-0.040740966796875,
0.00823211669921875,
0.0341796875,
-0.047943115234375,
0.0294036865234375,
-0.05010986328125,
-0.03619384765625,
0.018829345703125,
0.0204925537109375,
-0.068359375,
0.023590087890625,
-0.01038360595703125,
0.06610107421875,
-0.043243408203125,
0.05859375,
0.06549072265625,
-0.0357666015625,
-0.08172607421875,
-0.0000247955322265625,
0.00832366943359375,
-0.0709228515625,
0.050689697265625,
0.040313720703125,
0.0010309219360351562,
0.0073394775390625,
-0.0599365234375,
-0.048797607421875,
0.10205078125,
0.0269927978515625,
-0.0032062530517578125,
0.0249786376953125,
-0.0116119384765625,
0.00438690185546875,
-0.0377197265625,
0.041290283203125,
0.015411376953125,
0.02276611328125,
0.021575927734375,
-0.0609130859375,
0.0182037353515625,
-0.0300750732421875,
0.01515960693359375,
0.01509857177734375,
-0.06549072265625,
0.059967041015625,
-0.041168212890625,
-0.0087127685546875,
0.002918243408203125,
0.044097900390625,
0.0169219970703125,
0.023956298828125,
0.0350341796875,
0.0552978515625,
0.03619384765625,
-0.0178680419921875,
0.06671142578125,
0.0028171539306640625,
0.038604736328125,
0.0438232421875,
0.017333984375,
0.046173095703125,
0.026275634765625,
-0.01535797119140625,
0.031585693359375,
0.09222412109375,
-0.0228424072265625,
0.0206298828125,
0.01313018798828125,
-0.0027675628662109375,
0.00070953369140625,
0.006130218505859375,
-0.03460693359375,
0.05419921875,
0.0128021240234375,
-0.04254150390625,
-0.00806427001953125,
0.005157470703125,
0.006610870361328125,
-0.027069091796875,
-0.021453857421875,
0.032196044921875,
0.0012121200561523438,
-0.024627685546875,
0.07806396484375,
0.01763916015625,
0.06463623046875,
-0.01904296875,
0.00542449951171875,
-0.0211029052734375,
0.005401611328125,
-0.03338623046875,
-0.049560546875,
0.021728515625,
-0.0211944580078125,
-0.0029277801513671875,
0.007350921630859375,
0.051025390625,
-0.00966644287109375,
-0.0270538330078125,
0.010009765625,
0.0213623046875,
0.04052734375,
0.0020542144775390625,
-0.09124755859375,
0.0228729248046875,
0.01007080078125,
-0.046844482421875,
0.0205535888671875,
0.0229034423828125,
0.00457000732421875,
0.06378173828125,
0.04754638671875,
-0.0102996826171875,
0.01291656494140625,
-0.0234375,
0.0628662109375,
-0.05328369140625,
-0.017120361328125,
-0.06451416015625,
0.04559326171875,
-0.01454925537109375,
-0.0443115234375,
0.041168212890625,
0.052154541015625,
0.06317138671875,
0.002292633056640625,
0.039031982421875,
-0.0235595703125,
-0.004535675048828125,
-0.03826904296875,
0.049591064453125,
-0.0626220703125,
0.003009796142578125,
-0.006809234619140625,
-0.0518798828125,
-0.0288543701171875,
0.059722900390625,
-0.01837158203125,
0.0294647216796875,
0.042755126953125,
0.07861328125,
-0.032867431640625,
-0.01422882080078125,
0.007648468017578125,
-0.00045943260192871094,
-0.004886627197265625,
0.023284912109375,
0.032012939453125,
-0.06719970703125,
0.027435302734375,
-0.041656494140625,
-0.0157623291015625,
-0.018035888671875,
-0.056396484375,
-0.07733154296875,
-0.06573486328125,
-0.038787841796875,
-0.061248779296875,
-0.01715087890625,
0.07000732421875,
0.0859375,
-0.0433349609375,
-0.015838623046875,
0.00431060791015625,
0.0167236328125,
-0.011383056640625,
-0.016845703125,
0.04351806640625,
-0.0016269683837890625,
-0.050689697265625,
-0.0162506103515625,
-0.0022125244140625,
0.03485107421875,
0.01198577880859375,
-0.0156402587890625,
-0.0135650634765625,
-0.022918701171875,
0.024200439453125,
0.033355712890625,
-0.048675537109375,
-0.007457733154296875,
-0.018157958984375,
-0.0160675048828125,
0.0309600830078125,
0.042205810546875,
-0.03594970703125,
0.015594482421875,
0.016021728515625,
0.02606201171875,
0.0704345703125,
-0.0208740234375,
0.005977630615234375,
-0.059326171875,
0.048187255859375,
-0.012359619140625,
0.029083251953125,
0.028961181640625,
-0.022674560546875,
0.0452880859375,
0.028900146484375,
-0.032562255859375,
-0.06671142578125,
-0.00789642333984375,
-0.08441162109375,
-0.0026378631591796875,
0.075927734375,
-0.01983642578125,
-0.0450439453125,
0.028411865234375,
-0.0003483295440673828,
0.047943115234375,
-0.01049041748046875,
0.033294677734375,
0.01267242431640625,
-0.01551055908203125,
-0.050689697265625,
-0.055938720703125,
0.03533935546875,
0.0088958740234375,
-0.04693603515625,
-0.044158935546875,
-0.0038585662841796875,
0.050201416015625,
0.0161590576171875,
0.042999267578125,
-0.013397216796875,
0.0080413818359375,
0.005199432373046875,
0.038818359375,
-0.0279388427734375,
0.0007195472717285156,
-0.01568603515625,
-0.0015687942504882812,
-0.0159454345703125,
-0.054351806640625
]
] |
facebook/nougat-base | 2023-09-25T19:24:39.000Z | [
"transformers",
"pytorch",
"vision-encoder-decoder",
"vision",
"nougat",
"image-to-text",
"arxiv:2308.13418",
"license:apache-2.0",
"endpoints_compatible",
"has_space",
"region:us"
] | image-to-text | facebook | null | null | facebook/nougat-base | 54 | 7,239 | transformers | 2023-09-21T08:39:24 | ---
license: apache-2.0
tags:
- vision
- nougat
pipeline_tag: image-to-text
---
# Nougat model, base-sized version
Nougat model trained on PDF-to-markdown. It was introduced in the paper [Nougat: Neural Optical Understanding for Academic Documents](https://arxiv.org/abs/2308.13418) by Blecher et al. and first released in [this repository](https://github.com/facebookresearch/nougat/tree/main).
Disclaimer: The team releasing Nougat did not write a model card for this model so this model card has been written by the Hugging Face team.
Note: this model corresponds to the "0.1.0-base" version of the original repository.
## Model description
Nougat is a [Donut](https://huggingface.co/docs/transformers/model_doc/donut) model trained to transcribe scientific PDFs into an easy-to-use markdown format. The model consists of a Swin Transformer as vision encoder, and an mBART model as text decoder.
The model is trained to autoregressively predict the markdown given only the pixels of the PDF image as input.
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/nougat_architecture.jpg"
alt="drawing" width="600"/>
<small> Nougat high-level overview. Taken from the <a href="https://arxiv.org/abs/2308.13418">original paper</a>. </small>
## Intended uses & limitations
You can use the raw model for transcribing a PDF into Markdown. See the [model hub](https://huggingface.co/models?search=nougat) to look for other
fine-tuned versions that may interest you.
### How to use
We refer to the [docs](https://huggingface.co/docs/transformers/main/en/model_doc/nougat).
### BibTeX entry and citation info
```bibtex
@misc{blecher2023nougat,
title={Nougat: Neural Optical Understanding for Academic Documents},
author={Lukas Blecher and Guillem Cucurull and Thomas Scialom and Robert Stojnic},
year={2023},
eprint={2308.13418},
archivePrefix={arXiv},
primaryClass={cs.LG}
}
``` | 1,982 | [
[
-0.0285797119140625,
-0.0230865478515625,
0.0264129638671875,
0.01183319091796875,
-0.0233154296875,
-0.0264892578125,
0.0019273757934570312,
-0.037017822265625,
0.0149383544921875,
0.0474853515625,
-0.053436279296875,
-0.0220794677734375,
-0.05694580078125,
0.0182037353515625,
-0.0295562744140625,
0.08453369140625,
-0.0003216266632080078,
0.017181396484375,
0.0162506103515625,
-0.01377105712890625,
-0.0027408599853515625,
-0.017852783203125,
-0.043121337890625,
-0.02911376953125,
0.02056884765625,
0.00873565673828125,
0.04486083984375,
0.07354736328125,
0.04278564453125,
0.012115478515625,
-0.0234375,
-0.03570556640625,
-0.0101470947265625,
-0.03155517578125,
0.00984954833984375,
-0.03814697265625,
-0.06463623046875,
0.0103302001953125,
0.039764404296875,
0.044097900390625,
-0.005641937255859375,
0.004085540771484375,
0.006214141845703125,
0.042938232421875,
-0.0197296142578125,
0.015625,
-0.032257080078125,
0.01428985595703125,
0.006336212158203125,
-0.0003387928009033203,
-0.032623291015625,
-0.01690673828125,
0.01171875,
-0.0556640625,
0.035369873046875,
-0.015228271484375,
0.0869140625,
0.03570556640625,
-0.042388916015625,
-0.0118865966796875,
-0.037109375,
0.06329345703125,
-0.060699462890625,
0.05816650390625,
0.028961181640625,
0.024810791015625,
0.01274871826171875,
-0.045074462890625,
-0.038909912109375,
-0.0295257568359375,
-0.014892578125,
0.037109375,
-0.00962066650390625,
-0.00604248046875,
0.02056884765625,
0.056427001953125,
-0.05072021484375,
-0.001476287841796875,
-0.049041748046875,
-0.00327301025390625,
0.03759765625,
-0.0078887939453125,
0.0299072265625,
-0.0450439453125,
-0.04583740234375,
-0.021575927734375,
-0.0220184326171875,
-0.0279388427734375,
0.0167388916015625,
0.007049560546875,
-0.04290771484375,
0.042755126953125,
0.01251983642578125,
0.0494384765625,
0.027801513671875,
-0.00678253173828125,
0.0202484130859375,
-0.00933837890625,
-0.0305938720703125,
0.0091400146484375,
0.062744140625,
0.0283966064453125,
-0.0004916191101074219,
-0.0143890380859375,
-0.0197296142578125,
-0.0032215118408203125,
0.034698486328125,
-0.057861328125,
-0.032958984375,
-0.009552001953125,
-0.040313720703125,
-0.034912109375,
-0.01556396484375,
-0.05963134765625,
-0.032257080078125,
-0.034454345703125,
0.02252197265625,
-0.04833984375,
-0.03997802734375,
-0.0002465248107910156,
0.0028438568115234375,
0.0205078125,
0.050140380859375,
-0.09075927734375,
0.037353515625,
0.02142333984375,
0.060089111328125,
0.007434844970703125,
-0.021942138671875,
0.005039215087890625,
0.013275146484375,
-0.01446533203125,
0.052886962890625,
-0.005641937255859375,
-0.01541900634765625,
-0.02569580078125,
0.00984954833984375,
-0.0179443359375,
-0.033843994140625,
0.05169677734375,
-0.0301361083984375,
0.0157623291015625,
-0.00023889541625976562,
-0.029815673828125,
-0.037353515625,
0.03033447265625,
-0.0650634765625,
0.0850830078125,
0.0283355712890625,
-0.05609130859375,
0.02288818359375,
-0.06329345703125,
-0.019500732421875,
-0.003269195556640625,
0.0016298294067382812,
-0.0229644775390625,
0.0030841827392578125,
0.02056884765625,
0.02471923828125,
-0.0256805419921875,
-0.009124755859375,
-0.01666259765625,
-0.0028285980224609375,
-0.00479888916015625,
0.0015897750854492188,
0.06378173828125,
0.0113525390625,
-0.02276611328125,
0.005512237548828125,
-0.058990478515625,
-0.034576416015625,
0.032958984375,
-0.018798828125,
-0.00817108154296875,
-0.02886962890625,
0.0197296142578125,
0.017303466796875,
0.021453857421875,
-0.04718017578125,
0.03997802734375,
-0.003040313720703125,
0.05029296875,
0.039459228515625,
-0.0154266357421875,
0.06787109375,
-0.02447509765625,
0.02557373046875,
-0.0155792236328125,
0.036285400390625,
-0.025726318359375,
-0.0288238525390625,
-0.05450439453125,
-0.0297698974609375,
0.017974853515625,
0.045623779296875,
-0.032470703125,
0.0109100341796875,
-0.01372528076171875,
-0.061676025390625,
-0.03277587890625,
0.0012187957763671875,
0.01335906982421875,
0.061004638671875,
0.03997802734375,
-0.0191650390625,
-0.0302581787109375,
-0.06353759765625,
0.00362396240234375,
-0.0036334991455078125,
-0.01555633544921875,
0.0325927734375,
0.033416748046875,
-0.00860595703125,
0.044647216796875,
-0.039215087890625,
-0.039398193359375,
-0.014068603515625,
0.01495361328125,
0.0218505859375,
0.04119873046875,
0.049713134765625,
-0.07476806640625,
-0.02545166015625,
-0.00606536865234375,
-0.06591796875,
-0.0071258544921875,
-0.009613037109375,
-0.0031299591064453125,
0.03411865234375,
0.031585693359375,
-0.0343017578125,
0.0318603515625,
0.0384521484375,
-0.0081329345703125,
0.042694091796875,
0.0018329620361328125,
0.009490966796875,
-0.09173583984375,
0.0085601806640625,
0.0321044921875,
-0.00179290771484375,
-0.03759765625,
-0.004177093505859375,
0.0277252197265625,
-0.01128387451171875,
-0.03759765625,
0.040008544921875,
-0.04833984375,
-0.0014743804931640625,
-0.00197601318359375,
0.00605010986328125,
-0.0003523826599121094,
0.043548583984375,
0.033172607421875,
0.0272369384765625,
0.0236358642578125,
-0.03582763671875,
0.005306243896484375,
0.039886474609375,
0.0005826950073242188,
0.0513916015625,
-0.049896240234375,
0.00452423095703125,
0.0006060600280761719,
0.012725830078125,
-0.061676025390625,
-0.0091552734375,
0.03460693359375,
-0.032379150390625,
0.047637939453125,
-0.0433349609375,
-0.0518798828125,
-0.01531982421875,
-0.0095977783203125,
0.0214691162109375,
0.0458984375,
-0.0404052734375,
0.05328369140625,
0.01097869873046875,
-0.0007343292236328125,
-0.0244140625,
-0.07489013671875,
-0.0103302001953125,
0.006221771240234375,
-0.061309814453125,
0.0322265625,
-0.01151275634765625,
0.02862548828125,
0.01529693603515625,
-0.0260772705078125,
-0.0246124267578125,
-0.01593017578125,
0.0169677734375,
0.03094482421875,
-0.03045654296875,
-0.0082244873046875,
-0.0024204254150390625,
-0.02691650390625,
-0.019378662109375,
-0.01617431640625,
0.0304718017578125,
-0.0103759765625,
0.000652313232421875,
-0.03314208984375,
0.00409698486328125,
0.05340576171875,
-0.01415252685546875,
0.03521728515625,
0.047576904296875,
-0.035400390625,
0.00225830078125,
-0.0399169921875,
-0.032440185546875,
-0.032958984375,
0.005157470703125,
-0.0153961181640625,
-0.0491943359375,
0.046966552734375,
0.00504302978515625,
-0.002208709716796875,
0.048980712890625,
0.0245819091796875,
0.0146331787109375,
0.052581787109375,
0.07257080078125,
0.031585693359375,
0.057525634765625,
-0.0311431884765625,
0.013031005859375,
-0.0869140625,
-0.02606201171875,
-0.0212249755859375,
-0.006633758544921875,
-0.019805908203125,
-0.00560760498046875,
-0.001094818115234375,
0.0267486572265625,
-0.021820068359375,
0.0654296875,
-0.07879638671875,
0.0056610107421875,
0.031707763671875,
-0.000008881092071533203,
0.03277587890625,
-0.002132415771484375,
-0.0051116943359375,
-0.026214599609375,
-0.036468505859375,
-0.04779052734375,
0.06475830078125,
0.0203399658203125,
0.054351806640625,
0.009735107421875,
0.044708251953125,
-0.008544921875,
0.041656494140625,
-0.056427001953125,
0.040618896484375,
-0.010223388671875,
-0.06304931640625,
0.0038890838623046875,
-0.0301055908203125,
-0.0599365234375,
0.00757598876953125,
-0.0213775634765625,
-0.0755615234375,
0.0172119140625,
0.004978179931640625,
-0.00543975830078125,
0.04180908203125,
-0.0518798828125,
0.08123779296875,
-0.019195556640625,
-0.00769805908203125,
0.01953125,
-0.04107666015625,
0.038421630859375,
-0.006931304931640625,
-0.005771636962890625,
0.0023288726806640625,
-0.0110321044921875,
0.065185546875,
-0.0335693359375,
0.06109619140625,
0.0110321044921875,
-0.0034694671630859375,
0.0008106231689453125,
0.0157470703125,
0.00968170166015625,
-0.011505126953125,
-0.0211334228515625,
0.0413818359375,
0.019317626953125,
-0.035888671875,
-0.0458984375,
0.042572021484375,
-0.071044921875,
-0.00867462158203125,
-0.043853759765625,
-0.04168701171875,
0.02569580078125,
0.05242919921875,
0.0382080078125,
0.023590087890625,
-0.0028324127197265625,
-0.007732391357421875,
0.034942626953125,
-0.01171875,
0.0264892578125,
0.04913330078125,
-0.0194091796875,
-0.042755126953125,
0.054779052734375,
0.035888671875,
0.0162353515625,
0.030670166015625,
0.00968170166015625,
-0.0251922607421875,
-0.03558349609375,
-0.05816650390625,
0.03924560546875,
-0.048828125,
-0.01428985595703125,
-0.05596923828125,
-0.01727294921875,
-0.043548583984375,
-0.03240966796875,
-0.046783447265625,
-0.025146484375,
-0.034698486328125,
-0.00588226318359375,
0.04364013671875,
0.08026123046875,
-0.00878143310546875,
0.0411376953125,
-0.0634765625,
0.0236663818359375,
-0.003154754638671875,
0.040130615234375,
0.02423095703125,
-0.03094482421875,
-0.00920867919921875,
-0.01320648193359375,
-0.05035400390625,
-0.08489990234375,
0.04833984375,
-0.0206298828125,
0.03533935546875,
0.0201873779296875,
0.0090484619140625,
0.0278167724609375,
-0.0188140869140625,
0.061248779296875,
0.0257568359375,
-0.0657958984375,
0.03228759765625,
-0.0433349609375,
0.018280029296875,
0.03173828125,
0.038482666015625,
-0.030792236328125,
0.0030975341796875,
-0.06463623046875,
-0.065673828125,
0.07574462890625,
0.01247406005859375,
0.001842498779296875,
0.026702880859375,
0.0308380126953125,
0.017608642578125,
0.0220184326171875,
-0.056793212890625,
-0.01436614990234375,
-0.050506591796875,
-0.01361846923828125,
0.001495361328125,
-0.01247406005859375,
0.00606536865234375,
-0.02978515625,
0.041168212890625,
0.003688812255859375,
0.032684326171875,
0.0364990234375,
-0.01910400390625,
-0.00972747802734375,
0.004253387451171875,
0.04656982421875,
0.01416015625,
-0.0487060546875,
-0.0021152496337890625,
-0.004711151123046875,
-0.0614013671875,
-0.0132904052734375,
0.02398681640625,
-0.01413726806640625,
-0.00933837890625,
0.0203857421875,
0.054656982421875,
0.0287017822265625,
0.0008082389831542969,
0.03253173828125,
-0.0131683349609375,
-0.03289794921875,
-0.03173828125,
-0.0293731689453125,
0.003406524658203125,
0.00545501708984375,
0.039398193359375,
0.0279388427734375,
0.005657196044921875,
-0.01654052734375,
0.0197296142578125,
-0.007320404052734375,
-0.04766845703125,
-0.029937744140625,
0.07061767578125,
-0.00027632713317871094,
-0.017974853515625,
0.060699462890625,
-0.01468658447265625,
-0.033416748046875,
0.049407958984375,
0.046295166015625,
0.0635986328125,
-0.0121917724609375,
0.005809783935546875,
0.0408935546875,
0.049041748046875,
-0.002094268798828125,
0.0159759521484375,
-0.01409912109375,
-0.038787841796875,
-0.020660400390625,
-0.05206298828125,
-0.036163330078125,
0.019378662109375,
-0.062744140625,
0.03167724609375,
-0.060089111328125,
-0.00699615478515625,
0.01812744140625,
0.01111602783203125,
-0.040740966796875,
0.0197906494140625,
0.01319122314453125,
0.090087890625,
-0.06640625,
0.0543212890625,
0.06597900390625,
-0.06378173828125,
-0.076416015625,
-0.000965118408203125,
0.00879669189453125,
-0.040618896484375,
0.039306640625,
0.008514404296875,
0.0166168212890625,
0.01294708251953125,
-0.041473388671875,
-0.058197021484375,
0.0877685546875,
0.04833984375,
-0.0156707763671875,
0.0034942626953125,
-0.0178070068359375,
0.030853271484375,
-0.02490234375,
0.042755126953125,
0.00010228157043457031,
0.0469970703125,
0.031768798828125,
-0.0721435546875,
0.0283355712890625,
-0.041412353515625,
-0.01029205322265625,
0.01959228515625,
-0.056640625,
0.07574462890625,
-0.0138092041015625,
-0.009735107421875,
0.01654052734375,
0.03253173828125,
-0.0060272216796875,
0.0256500244140625,
0.023834228515625,
0.058502197265625,
0.0355224609375,
-0.0095062255859375,
0.10986328125,
-0.01190185546875,
0.0350341796875,
0.061798095703125,
0.0016489028930664062,
0.045135498046875,
0.03326416015625,
0.0032501220703125,
0.022064208984375,
0.060394287109375,
-0.042083740234375,
0.04022216796875,
-0.0136566162109375,
0.0181732177734375,
-0.007503509521484375,
-0.004772186279296875,
-0.0267486572265625,
0.0003457069396972656,
0.00637054443359375,
-0.04296875,
-0.0245819091796875,
-0.00818634033203125,
0.017181396484375,
-0.009033203125,
-0.02691650390625,
0.054656982421875,
0.01812744140625,
-0.0243072509765625,
0.062225341796875,
-0.0099334716796875,
0.056549072265625,
-0.042266845703125,
-0.0007987022399902344,
0.0015964508056640625,
0.0167694091796875,
-0.044464111328125,
-0.049774169921875,
0.033721923828125,
-0.01544189453125,
-0.042083740234375,
-0.0147247314453125,
0.0305023193359375,
-0.01110076904296875,
-0.03863525390625,
0.04254150390625,
0.0108184814453125,
0.031036376953125,
0.00211334228515625,
-0.064697265625,
0.019073486328125,
-0.00962066650390625,
-0.0311126708984375,
0.02734375,
0.00914764404296875,
-0.0020160675048828125,
0.0330810546875,
0.0634765625,
-0.01486968994140625,
0.019134521484375,
-0.01178741455078125,
0.0570068359375,
-0.039215087890625,
-0.040557861328125,
-0.0467529296875,
0.042083740234375,
-0.0049285888671875,
-0.037261962890625,
0.0672607421875,
0.043365478515625,
0.061248779296875,
-0.018768310546875,
0.048614501953125,
-0.00513458251953125,
0.01322174072265625,
-0.0164794921875,
0.0762939453125,
-0.07708740234375,
-0.0156402587890625,
-0.044708251953125,
-0.07830810546875,
-0.052520751953125,
0.06463623046875,
-0.048431396484375,
0.00502777099609375,
0.051605224609375,
0.05010986328125,
-0.03955078125,
0.0237884521484375,
0.01416778564453125,
0.026702880859375,
0.0024471282958984375,
0.014434814453125,
0.0272216796875,
-0.06280517578125,
0.0469970703125,
-0.040740966796875,
-0.0250091552734375,
-0.037109375,
-0.07061767578125,
-0.057769775390625,
-0.051666259765625,
-0.042266845703125,
-0.03765869140625,
-0.0170135498046875,
0.0231170654296875,
0.0804443359375,
-0.041961669921875,
-0.01064300537109375,
-0.00701904296875,
0.00534820556640625,
-0.01593017578125,
-0.0164794921875,
0.03961181640625,
0.0160980224609375,
-0.052825927734375,
0.004856109619140625,
0.01369476318359375,
0.033599853515625,
0.0029964447021484375,
-0.0151214599609375,
-0.0164794921875,
0.00975799560546875,
0.0283966064453125,
0.040618896484375,
-0.033477783203125,
0.010284423828125,
0.02618408203125,
-0.0215911865234375,
0.0287017822265625,
0.02789306640625,
-0.022003173828125,
0.0072479248046875,
0.03759765625,
0.020477294921875,
0.07037353515625,
-0.026123046875,
0.0287017822265625,
-0.024322509765625,
0.0165863037109375,
0.0115203857421875,
0.037109375,
0.053131103515625,
-0.031829833984375,
0.052490234375,
0.01531982421875,
-0.038543701171875,
-0.06939697265625,
0.0207366943359375,
-0.12200927734375,
-0.005146026611328125,
0.07098388671875,
0.008819580078125,
-0.053741455078125,
0.019775390625,
-0.03387451171875,
0.0394287109375,
-0.056182861328125,
0.035369873046875,
0.026824951171875,
-0.00664520263671875,
-0.0692138671875,
-0.049896240234375,
0.01505279541015625,
0.01242828369140625,
-0.02459716796875,
-0.00897216796875,
0.0096588134765625,
0.044647216796875,
0.0423583984375,
0.0423583984375,
-0.041748046875,
0.016448974609375,
0.01751708984375,
0.045989990234375,
-0.0237884521484375,
0.0042877197265625,
-0.044891357421875,
-0.0028400421142578125,
-0.01020050048828125,
-0.0022144317626953125
]
] |
oliverguhr/fullstop-punctuation-multilingual-base | 2023-03-21T09:16:18.000Z | [
"transformers",
"pytorch",
"tensorboard",
"safetensors",
"xlm-roberta",
"token-classification",
"punctuation prediction",
"punctuation",
"en",
"de",
"fr",
"it",
"nl",
"multilingual",
"dataset:wmt/europarl",
"arxiv:2301.03319",
"license:mit",
"autotrain_compatible",
"endpoints_compatible",
"region:us"
] | token-classification | oliverguhr | null | null | oliverguhr/fullstop-punctuation-multilingual-base | 6 | 7,234 | transformers | 2022-03-22T09:03:02 | ---
language:
- en
- de
- fr
- it
- nl
- multilingual
tags:
- punctuation prediction
- punctuation
datasets: wmt/europarl
license: mit
widget:
- text: "Ondanks dat het nu bijna voorjaar is hebben we nog steds best koude dagen"
example_title: "Dutch"
- text: "Ho sentito che ti sei laureata il che mi fa molto piacere"
example_title: "Italian"
- text: "Tous les matins vers quatre heures mon père ouvrait la porte de ma chambre"
example_title: "French"
- text: "Ist das eine Frage Frau Müller"
example_title: "German"
- text: "My name is Clara and I live in Berkeley California"
example_title: "English"
metrics:
- f1
---
# Work in progress
## Classification report over all languages
```
precision recall f1-score support
0 0.99 0.99 0.99 47903344
. 0.94 0.95 0.95 2798780
, 0.85 0.84 0.85 3451618
? 0.88 0.85 0.87 88876
- 0.61 0.32 0.42 157863
: 0.72 0.52 0.60 103789
accuracy 0.98 54504270
macro avg 0.83 0.75 0.78 54504270
weighted avg 0.98 0.98 0.98 54504270
```
## How to cite us
```
@article{guhr-EtAl:2021:fullstop,
title={FullStop: Multilingual Deep Models for Punctuation Prediction},
author = {Guhr, Oliver and Schumann, Anne-Kathrin and Bahrmann, Frank and Böhme, Hans Joachim},
booktitle = {Proceedings of the Swiss Text Analytics Conference 2021},
month = {June},
year = {2021},
address = {Winterthur, Switzerland},
publisher = {CEUR Workshop Proceedings},
url = {http://ceur-ws.org/Vol-2957/sepp_paper4.pdf}
}
```
```
@misc{https://doi.org/10.48550/arxiv.2301.03319,
doi = {10.48550/ARXIV.2301.03319},
url = {https://arxiv.org/abs/2301.03319},
author = {Vandeghinste, Vincent and Guhr, Oliver},
keywords = {Computation and Language (cs.CL), Artificial Intelligence (cs.AI), FOS: Computer and information sciences, FOS: Computer and information sciences, I.2.7},
title = {FullStop:Punctuation and Segmentation Prediction for Dutch with Transformers},
publisher = {arXiv},
year = {2023},
copyright = {Creative Commons Attribution Share Alike 4.0 International}
}
```
| 2,355 | [
[
-0.012420654296875,
-0.05426025390625,
0.045623779296875,
0.041595458984375,
-0.004917144775390625,
0.01123809814453125,
-0.044952392578125,
-0.04388427734375,
-0.00165557861328125,
0.0194854736328125,
-0.024658203125,
-0.0667724609375,
-0.03887939453125,
0.03955078125,
-0.0038967132568359375,
0.040924072265625,
-0.016326904296875,
0.0032901763916015625,
-0.0203704833984375,
-0.0157928466796875,
-0.026153564453125,
-0.0748291015625,
-0.034393310546875,
-0.0119476318359375,
0.0275115966796875,
0.03289794921875,
0.0204620361328125,
0.044647216796875,
0.0369873046875,
0.0280609130859375,
-0.01277923583984375,
0.0159149169921875,
0.00673675537109375,
-0.006000518798828125,
0.0015306472778320312,
-0.0212249755859375,
-0.032379150390625,
-0.0026702880859375,
0.058990478515625,
0.060882568359375,
-0.006130218505859375,
-0.007770538330078125,
0.008636474609375,
0.049285888671875,
-0.0328369140625,
0.032623291015625,
-0.0233001708984375,
-0.005344390869140625,
-0.0211639404296875,
0.00576019287109375,
-0.043182373046875,
-0.03472900390625,
-0.007312774658203125,
-0.041473388671875,
0.00373077392578125,
0.0157470703125,
0.06378173828125,
0.0002313852310180664,
-0.011322021484375,
-0.011138916015625,
-0.043121337890625,
0.07421875,
-0.052032470703125,
0.04144287109375,
0.025848388671875,
-0.00345611572265625,
-0.0013904571533203125,
-0.059600830078125,
-0.0516357421875,
-0.0106964111328125,
-0.02056884765625,
0.00965118408203125,
-0.017669677734375,
-0.0012903213500976562,
0.02447509765625,
0.0292205810546875,
-0.046417236328125,
0.0093841552734375,
-0.0284576416015625,
-0.028533935546875,
0.044769287109375,
0.01019287109375,
0.003963470458984375,
0.0006933212280273438,
-0.0132904052734375,
-0.0251617431640625,
-0.0284423828125,
0.0166168212890625,
0.0201568603515625,
0.0189971923828125,
-0.040130615234375,
0.031829833984375,
-0.0163116455078125,
0.047882080078125,
-0.01074981689453125,
-0.001987457275390625,
0.056793212890625,
-0.04608154296875,
-0.01165771484375,
-0.0139923095703125,
0.0987548828125,
0.0014276504516601562,
0.043182373046875,
0.00714111328125,
0.0140533447265625,
-0.004558563232421875,
-0.0100250244140625,
-0.05572509765625,
-0.03009033203125,
0.0007486343383789062,
-0.0211029052734375,
-0.005886077880859375,
0.030609130859375,
-0.0576171875,
-0.007778167724609375,
-0.01849365234375,
-0.01137542724609375,
-0.05377197265625,
-0.007778167724609375,
0.024078369140625,
0.01010894775390625,
0.001331329345703125,
0.02301025390625,
-0.03997802734375,
0.01187896728515625,
0.022918701171875,
0.048065185546875,
-0.0094146728515625,
-0.04736328125,
-0.0247650146484375,
-0.0117950439453125,
-0.00800323486328125,
0.0653076171875,
-0.04132080078125,
-0.02642822265625,
0.00998687744140625,
0.0198211669921875,
-0.042724609375,
-0.030670166015625,
0.0789794921875,
-0.01477813720703125,
0.04052734375,
-0.0159149169921875,
-0.04974365234375,
-0.03546142578125,
0.018707275390625,
-0.060089111328125,
0.0906982421875,
-0.01120758056640625,
-0.055938720703125,
0.00537872314453125,
-0.049591064453125,
-0.049407958984375,
-0.01056671142578125,
-0.005130767822265625,
-0.0537109375,
-0.01236724853515625,
0.0286407470703125,
0.0302886962890625,
-0.0234527587890625,
0.0217132568359375,
-0.0157470703125,
-0.01503753662109375,
0.029754638671875,
-0.035247802734375,
0.07110595703125,
0.00992584228515625,
-0.021453857421875,
-0.003849029541015625,
-0.06329345703125,
0.00699615478515625,
0.0022296905517578125,
-0.04144287109375,
-0.01190185546875,
-0.005893707275390625,
0.01422882080078125,
0.0221099853515625,
0.0276336669921875,
-0.036712646484375,
0.007678985595703125,
-0.03326416015625,
0.030120849609375,
0.0343017578125,
0.00226593017578125,
0.030548095703125,
-0.006809234619140625,
0.040924072265625,
0.00798797607421875,
0.004009246826171875,
-0.024261474609375,
-0.037200927734375,
-0.06317138671875,
-0.03155517578125,
0.04571533203125,
0.07647705078125,
-0.042510986328125,
0.06597900390625,
-0.022857666015625,
-0.03857421875,
-0.0205230712890625,
0.00750732421875,
0.051361083984375,
0.0413818359375,
0.04144287109375,
-0.01474761962890625,
-0.057647705078125,
-0.057586669921875,
-0.0195770263671875,
-0.0296173095703125,
0.01395416259765625,
0.01448822021484375,
0.04595947265625,
-0.0054168701171875,
0.057342529296875,
-0.04290771484375,
-0.0197601318359375,
-0.012054443359375,
0.01419830322265625,
0.033477783203125,
0.04815673828125,
0.036865234375,
-0.053314208984375,
-0.057769775390625,
0.002445220947265625,
-0.050262451171875,
-0.007419586181640625,
0.005374908447265625,
-0.0066986083984375,
0.04034423828125,
0.045562744140625,
-0.0343017578125,
0.0250396728515625,
0.0162506103515625,
-0.0362548828125,
0.058929443359375,
0.0005321502685546875,
0.0205230712890625,
-0.09063720703125,
0.040496826171875,
0.006282806396484375,
-0.001567840576171875,
-0.03448486328125,
-0.02374267578125,
0.0199737548828125,
0.01006317138671875,
-0.038970947265625,
0.05548095703125,
-0.0399169921875,
0.0278778076171875,
0.0250396728515625,
-0.01331329345703125,
0.00891876220703125,
0.056396484375,
0.0150146484375,
0.061431884765625,
0.034454345703125,
-0.042388916015625,
-0.0025539398193359375,
0.034454345703125,
-0.04327392578125,
0.0284423828125,
-0.042877197265625,
0.000732421875,
0.028594970703125,
0.0026531219482421875,
-0.0587158203125,
0.00371551513671875,
0.0215301513671875,
-0.05963134765625,
0.03094482421875,
0.00009387731552124023,
-0.0626220703125,
-0.0277099609375,
-0.019256591796875,
0.034912109375,
0.02227783203125,
-0.012786865234375,
0.034576416015625,
0.01337432861328125,
-0.0272064208984375,
-0.042816162109375,
-0.057098388671875,
0.0006394386291503906,
-0.005420684814453125,
-0.047882080078125,
0.0271148681640625,
-0.035003662109375,
-0.01187896728515625,
-0.00980377197265625,
0.016082763671875,
-0.005489349365234375,
0.01035308837890625,
0.0014753341674804688,
0.02105712890625,
-0.01393890380859375,
0.0028095245361328125,
-0.008270263671875,
-0.01322174072265625,
-0.009246826171875,
-0.019683837890625,
0.045989990234375,
-0.019378662109375,
-0.022674560546875,
-0.03240966796875,
0.02752685546875,
0.0341796875,
-0.0298004150390625,
0.04144287109375,
0.053558349609375,
-0.026123046875,
0.016357421875,
-0.046417236328125,
0.00539398193359375,
-0.034698486328125,
0.03155517578125,
-0.029083251953125,
-0.07269287109375,
0.06103515625,
0.0131378173828125,
-0.0150909423828125,
0.058258056640625,
0.0445556640625,
-0.0008039474487304688,
0.059295654296875,
0.06109619140625,
-0.0152740478515625,
0.042724609375,
-0.0169830322265625,
0.013427734375,
-0.047332763671875,
-0.0347900390625,
-0.06561279296875,
0.0025386810302734375,
-0.047210693359375,
-0.0300445556640625,
0.0122833251953125,
0.031158447265625,
-0.0200347900390625,
0.0335693359375,
-0.03363037109375,
0.031402587890625,
0.04864501953125,
-0.0212554931640625,
0.024261474609375,
0.029876708984375,
-0.05584716796875,
-0.0136871337890625,
-0.040130615234375,
-0.02886962890625,
0.06072998046875,
0.01171875,
0.0215911865234375,
-0.0024356842041015625,
0.044769287109375,
0.01076507568359375,
-0.007541656494140625,
-0.0531005859375,
0.042327880859375,
-0.021514892578125,
-0.02801513671875,
-0.0188140869140625,
-0.02984619140625,
-0.080810546875,
0.00423431396484375,
-0.0012693405151367188,
-0.0626220703125,
0.0259552001953125,
-0.00937652587890625,
-0.027740478515625,
0.020965576171875,
-0.0562744140625,
0.0689697265625,
0.0029735565185546875,
-0.0093841552734375,
0.00040602684020996094,
-0.046875,
0.00209808349609375,
-0.00274658203125,
0.036163330078125,
-0.0074462890625,
0.0145416259765625,
0.0675048828125,
-0.0240631103515625,
0.07012939453125,
0.002681732177734375,
-0.0005025863647460938,
0.028656005859375,
0.01001739501953125,
0.039794921875,
0.006072998046875,
-0.0205230712890625,
0.047515869140625,
-0.00797271728515625,
-0.0200347900390625,
-0.01480865478515625,
0.0404052734375,
-0.050811767578125,
-0.01094818115234375,
-0.0533447265625,
-0.034912109375,
-0.006908416748046875,
0.036102294921875,
0.0230865478515625,
0.0221099853515625,
-0.03582763671875,
0.00862884521484375,
0.0357666015625,
-0.0200958251953125,
0.059783935546875,
0.055389404296875,
0.01520538330078125,
-0.06988525390625,
0.058013916015625,
0.00812530517578125,
-0.01462554931640625,
0.05584716796875,
-0.00485992431640625,
-0.017608642578125,
-0.030364990234375,
-0.0269775390625,
0.03173828125,
-0.04925537109375,
-0.02386474609375,
-0.058258056640625,
-0.021820068359375,
-0.040771484375,
-0.0120849609375,
-0.0118408203125,
-0.047454833984375,
-0.04583740234375,
-0.0068817138671875,
0.030181884765625,
0.0273590087890625,
-0.01497650146484375,
0.014862060546875,
-0.05853271484375,
-0.0176849365234375,
-0.007480621337890625,
0.033203125,
-0.01251220703125,
-0.039825439453125,
-0.048065185546875,
-0.01238250732421875,
-0.0305938720703125,
-0.04437255859375,
0.043121337890625,
0.0233306884765625,
0.05096435546875,
0.007015228271484375,
-0.00891876220703125,
0.04693603515625,
-0.045806884765625,
0.06494140625,
0.031036376953125,
-0.07550048828125,
0.032379150390625,
-0.01275634765625,
0.028778076171875,
0.0579833984375,
0.012542724609375,
-0.058563232421875,
-0.021453857421875,
-0.057342529296875,
-0.069091796875,
0.058197021484375,
0.02587890625,
-0.01074981689453125,
-0.014434814453125,
0.00806427001953125,
0.00835418701171875,
0.01244354248046875,
-0.05084228515625,
-0.0335693359375,
0.01125335693359375,
-0.046295166015625,
-0.0175628662109375,
-0.027191162109375,
0.006206512451171875,
-0.03900146484375,
0.0556640625,
-0.013275146484375,
0.03546142578125,
0.01995849609375,
-0.024627685546875,
0.008209228515625,
0.0269317626953125,
0.0689697265625,
0.042449951171875,
-0.0036773681640625,
0.0016679763793945312,
0.0129852294921875,
-0.050811767578125,
-0.003387451171875,
0.036376953125,
-0.004688262939453125,
0.01277923583984375,
0.05084228515625,
0.05572509765625,
0.0170745849609375,
-0.03546142578125,
0.035186767578125,
0.005397796630859375,
-0.0333251953125,
-0.0533447265625,
-0.0239410400390625,
-0.006855010986328125,
0.0157928466796875,
0.0272674560546875,
0.0176239013671875,
0.0009312629699707031,
-0.004802703857421875,
0.01094818115234375,
0.004207611083984375,
-0.032745361328125,
-0.04388427734375,
0.03253173828125,
0.0117340087890625,
-0.01464080810546875,
0.04925537109375,
-0.0255889892578125,
-0.059783935546875,
0.04937744140625,
0.0238037109375,
0.07391357421875,
-0.021697998046875,
-0.004913330078125,
0.06427001953125,
0.0325927734375,
-0.010009765625,
0.0457763671875,
0.005725860595703125,
-0.054534912109375,
-0.0428466796875,
-0.060699462890625,
-0.017364501953125,
0.0024318695068359375,
-0.0302886962890625,
0.0294036865234375,
-0.0243682861328125,
-0.0015039443969726562,
0.007114410400390625,
0.0203857421875,
-0.0645751953125,
0.0093841552734375,
0.0186920166015625,
0.06854248046875,
-0.06646728515625,
0.07269287109375,
0.0689697265625,
-0.0567626953125,
-0.05084228515625,
-0.01393890380859375,
-0.01071929931640625,
-0.049468994140625,
0.033477783203125,
0.01140594482421875,
0.01155853271484375,
-0.0019855499267578125,
-0.03887939453125,
-0.07684326171875,
0.0772705078125,
0.027496337890625,
-0.05303955078125,
0.003475189208984375,
0.0105133056640625,
0.03253173828125,
-0.0175018310546875,
0.0207061767578125,
0.042327880859375,
0.059051513671875,
-0.008331298828125,
-0.09393310546875,
0.00048613548278808594,
-0.047576904296875,
0.0007662773132324219,
0.0114288330078125,
-0.050018310546875,
0.0673828125,
-0.01082611083984375,
-0.0200653076171875,
0.0189056396484375,
0.05352783203125,
0.0281982421875,
0.015655517578125,
0.0178070068359375,
0.061431884765625,
0.06353759765625,
-0.0218505859375,
0.06451416015625,
-0.0340576171875,
0.026702880859375,
0.07684326171875,
0.0006194114685058594,
0.058135986328125,
0.03350830078125,
-0.03717041015625,
0.07525634765625,
0.05072021484375,
-0.018890380859375,
0.0307464599609375,
-0.00685882568359375,
0.008056640625,
-0.0239715576171875,
0.005016326904296875,
-0.0298919677734375,
0.0176544189453125,
0.028076171875,
-0.044647216796875,
0.01232147216796875,
-0.00713348388671875,
0.0257568359375,
0.016937255859375,
0.00302886962890625,
0.05499267578125,
0.0127105712890625,
-0.042205810546875,
0.05352783203125,
0.01480865478515625,
0.0489501953125,
-0.0280609130859375,
-0.0082855224609375,
-0.02032470703125,
0.02301025390625,
-0.005123138427734375,
-0.0604248046875,
0.006114959716796875,
-0.01105499267578125,
-0.049957275390625,
-0.007381439208984375,
0.042816162109375,
-0.03802490234375,
-0.06878662109375,
0.024139404296875,
0.042572021484375,
0.00849151611328125,
-0.007579803466796875,
-0.05963134765625,
0.0095062255859375,
0.01428985595703125,
-0.0251312255859375,
-0.0001195669174194336,
0.046417236328125,
-0.0179290771484375,
0.032501220703125,
0.047149658203125,
-0.00566864013671875,
0.00225067138671875,
0.01030731201171875,
0.05010986328125,
-0.06390380859375,
-0.0533447265625,
-0.07232666015625,
0.056396484375,
-0.0194549560546875,
-0.03192138671875,
0.053985595703125,
0.068603515625,
0.07269287109375,
-0.00632476806640625,
0.07147216796875,
-0.0137481689453125,
0.047882080078125,
-0.03802490234375,
0.043121337890625,
-0.0299530029296875,
0.0023899078369140625,
-0.0269012451171875,
-0.06854248046875,
-0.0513916015625,
0.035888671875,
-0.0239410400390625,
0.0033512115478515625,
0.0894775390625,
0.043182373046875,
0.002796173095703125,
-0.019989013671875,
0.01482391357421875,
0.021942138671875,
0.0021419525146484375,
0.0498046875,
0.0233306884765625,
-0.04156494140625,
0.052703857421875,
-0.031585693359375,
0.0002073049545288086,
-0.020233154296875,
-0.06158447265625,
-0.06036376953125,
-0.06622314453125,
-0.04815673828125,
-0.024658203125,
-0.0092926025390625,
0.05560302734375,
0.03411865234375,
-0.09002685546875,
-0.0261077880859375,
-0.00313568115234375,
0.0171661376953125,
-0.0300750732421875,
-0.0165863037109375,
0.053253173828125,
-0.02301025390625,
-0.0635986328125,
0.03399658203125,
0.005855560302734375,
0.0121002197265625,
0.00875091552734375,
-0.0192718505859375,
-0.043121337890625,
-0.0113372802734375,
0.037109375,
0.03717041015625,
-0.058441162109375,
0.0009379386901855469,
0.0159759521484375,
-0.020965576171875,
0.021453857421875,
0.041412353515625,
-0.044769287109375,
0.018157958984375,
0.03997802734375,
0.0223388671875,
0.0285491943359375,
-0.0189971923828125,
0.043121337890625,
-0.0704345703125,
0.0255584716796875,
0.0172271728515625,
0.04278564453125,
0.04058837890625,
-0.0180511474609375,
0.057159423828125,
0.042236328125,
-0.03765869140625,
-0.05035400390625,
0.002079010009765625,
-0.07318115234375,
-0.01486968994140625,
0.08984375,
-0.013885498046875,
-0.01280975341796875,
-0.003570556640625,
-0.004314422607421875,
0.03173828125,
-0.046875,
0.0338134765625,
0.060516357421875,
0.021026611328125,
-0.0143585205078125,
-0.0227813720703125,
0.038482666015625,
0.033172607421875,
-0.0538330078125,
0.0023059844970703125,
0.040130615234375,
0.0209197998046875,
0.03448486328125,
0.03997802734375,
-0.0185546875,
0.013458251953125,
-0.0196075439453125,
0.055511474609375,
-0.018707275390625,
-0.018310546875,
-0.0328369140625,
0.004978179931640625,
-0.007511138916015625,
-0.004726409912109375
]
] |
medalpaca/medalpaca-7b | 2023-07-18T21:54:12.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"medical",
"en",
"arxiv:2303.14070",
"license:cc",
"endpoints_compatible",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | medalpaca | null | null | medalpaca/medalpaca-7b | 35 | 7,221 | transformers | 2023-03-29T17:54:49 | ---
license: cc
language:
- en
library_name: transformers
pipeline_tag: text-generation
tags:
- medical
---
# MedAlpaca 7b
## Table of Contents
[Model Description](#model-description)
- [Architecture](#architecture)
- [Training Data](#trainig-data)
[Model Usage](#model-usage)
[Limitations](#limitations)
## Model Description
### Architecture
`medalpaca-7b` is a large language model specifically fine-tuned for medical domain tasks.
It is based on LLaMA (Large Language Model Meta AI) and contains 7 billion parameters.
The primary goal of this model is to improve question-answering and medical dialogue tasks.
Architecture
### Training Data
The training data for this project was sourced from various resources.
Firstly, we used Anki flashcards to automatically generate questions,
from the front of the cards and anwers from the back of the card.
Secondly, we generated medical question-answer pairs from [Wikidoc](https://www.wikidoc.org/index.php/Main_Page).
We extracted paragraphs with relevant headings, and used Chat-GPT 3.5
to generate questions from the headings and using the corresponding paragraphs
as answers. This dataset is still under development and we believe
that approximately 70% of these question answer pairs are factual correct.
Thirdly, we used StackExchange to extract question-answer pairs, taking the
top-rated question from five categories: Academia, Bioinformatics, Biology,
Fitness, and Health. Additionally, we used a dataset from [ChatDoctor](https://arxiv.org/abs/2303.14070)
consisting of 200,000 question-answer pairs, available at https://github.com/Kent0n-Li/ChatDoctor.
| Source | n items |
|------------------------------|--------|
| ChatDoc large | 200000 |
| wikidoc | 67704 |
| Stackexchange academia | 40865 |
| Anki flashcards | 33955 |
| Stackexchange biology | 27887 |
| Stackexchange fitness | 9833 |
| Stackexchange health | 7721 |
| Wikidoc patient information | 5942 |
| Stackexchange bioinformatics | 5407 |
## Model Usage
To evaluate the performance of the model on a specific dataset, you can use the Hugging Face Transformers library's built-in evaluation scripts. Please refer to the evaluation guide for more information.
Inference
You can use the model for inference tasks like question-answering and medical dialogues using the Hugging Face Transformers library. Here's an example of how to use the model for a question-answering task:
```python
from transformers import pipeline
pl = pipeline("text-generation", model="medalpaca/medalpaca-7b", tokenizer="medalpaca/medalpaca-7b")
question = "What are the symptoms of diabetes?"
context = "Diabetes is a metabolic disease that causes high blood sugar. The symptoms include increased thirst, frequent urination, and unexplained weight loss."
answer = pl(f"Context: {context}\n\nQuestion: {question}\n\nAnswer: ")
print(answer)
```
## Limitations
The model may not perform effectively outside the scope of the medical domain.
The training data primarily targets the knowledge level of medical students,
which may result in limitations when addressing the needs of board-certified physicians.
The model has not been tested in real-world applications, so its efficacy and accuracy are currently unknown.
It should never be used as a substitute for a doctor's opinion and must be treated as a research tool only. | 3,470 | [
[
-0.030242919921875,
-0.07421875,
0.045623779296875,
0.006450653076171875,
-0.01113128662109375,
-0.015533447265625,
0.0120849609375,
-0.019073486328125,
0.020751953125,
0.045013427734375,
-0.042816162109375,
-0.034088134765625,
-0.050140380859375,
0.016357421875,
-0.00952911376953125,
0.090087890625,
0.004734039306640625,
0.0215301513671875,
-0.0418701171875,
-0.01416015625,
-0.032623291015625,
-0.041961669921875,
-0.04986572265625,
-0.038970947265625,
0.040985107421875,
0.0264434814453125,
0.04168701171875,
0.04022216796875,
0.041534423828125,
0.01971435546875,
-0.01556396484375,
0.00726318359375,
-0.0255279541015625,
-0.01235198974609375,
-0.007183074951171875,
-0.054168701171875,
-0.042236328125,
0.0025882720947265625,
0.01464080810546875,
0.062255859375,
-0.00203704833984375,
0.038787841796875,
-0.01561737060546875,
0.041534423828125,
-0.0227813720703125,
0.031280517578125,
-0.019012451171875,
-0.001247406005859375,
-0.0009675025939941406,
-0.0084686279296875,
-0.008544921875,
-0.02960205078125,
0.0092620849609375,
-0.02978515625,
0.00682830810546875,
0.004016876220703125,
0.06854248046875,
0.025970458984375,
-0.023651123046875,
-0.0286865234375,
-0.021148681640625,
0.036529541015625,
-0.07244873046875,
0.0076904296875,
0.050048828125,
0.023468017578125,
0.0016927719116210938,
-0.053070068359375,
-0.059906005859375,
-0.0146331787109375,
-0.013946533203125,
0.021881103515625,
-0.0010557174682617188,
0.00478363037109375,
0.03912353515625,
0.026519775390625,
-0.060455322265625,
-0.01007080078125,
-0.0535888671875,
-0.001678466796875,
0.048583984375,
0.02166748046875,
0.03729248046875,
-0.0384521484375,
-0.0168304443359375,
-0.010101318359375,
-0.033966064453125,
0.0174713134765625,
0.006381988525390625,
-0.0048065185546875,
-0.0240631103515625,
0.056915283203125,
-0.027587890625,
0.03948974609375,
0.0146636962890625,
-0.021759033203125,
0.050323486328125,
-0.012786865234375,
-0.0219573974609375,
-0.01103973388671875,
0.06768798828125,
0.034149169921875,
0.0067596435546875,
-0.01165771484375,
0.00957489013671875,
0.01611328125,
0.0309600830078125,
-0.06878662109375,
-0.00974273681640625,
0.054168701171875,
-0.043182373046875,
-0.027313232421875,
-0.01428985595703125,
-0.058441162109375,
-0.024139404296875,
-0.01049041748046875,
0.035736083984375,
-0.0299530029296875,
-0.0096588134765625,
0.0103759765625,
-0.007080078125,
0.0231781005859375,
0.00812530517578125,
-0.056365966796875,
0.0191497802734375,
0.0361328125,
0.05096435546875,
-0.01184844970703125,
-0.01666259765625,
-0.02630615234375,
-0.0173492431640625,
-0.015625,
0.05718994140625,
-0.0219573974609375,
-0.0205230712890625,
-0.0158538818359375,
0.027313232421875,
-0.0269775390625,
-0.051513671875,
0.04931640625,
-0.032073974609375,
0.032684326171875,
-0.03857421875,
-0.05340576171875,
-0.023895263671875,
0.0283203125,
-0.045013427734375,
0.069580078125,
0.019500732421875,
-0.07177734375,
-0.003002166748046875,
-0.05780029296875,
-0.017822265625,
-0.00682830810546875,
-0.01345062255859375,
-0.04046630859375,
-0.0096435546875,
0.041229248046875,
0.04278564453125,
-0.042144775390625,
0.036468505859375,
-0.019073486328125,
-0.0157623291015625,
0.037567138671875,
-0.01390838623046875,
0.0628662109375,
0.016387939453125,
-0.019378662109375,
-0.0007805824279785156,
-0.065185546875,
-0.012359619140625,
0.0167999267578125,
-0.017242431640625,
-0.0012235641479492188,
-0.01139068603515625,
-0.0183258056640625,
0.02850341796875,
0.017547607421875,
-0.05303955078125,
0.015167236328125,
-0.0433349609375,
0.033050537109375,
0.0270843505859375,
0.018829345703125,
0.00867462158203125,
-0.04925537109375,
0.059112548828125,
0.013885498046875,
0.0172882080078125,
-0.0072174072265625,
-0.0640869140625,
-0.054901123046875,
-0.007114410400390625,
0.023040771484375,
0.0615234375,
-0.0655517578125,
0.026031494140625,
-0.00618743896484375,
-0.0506591796875,
-0.06890869140625,
-0.0008816719055175781,
0.040283203125,
0.08526611328125,
0.045989990234375,
-0.026092529296875,
-0.033477783203125,
-0.07379150390625,
0.01207733154296875,
-0.03009033203125,
-0.0007691383361816406,
0.037109375,
0.049591064453125,
-0.0169677734375,
0.04876708984375,
-0.05035400390625,
0.00455474853515625,
-0.0288238525390625,
-0.0029144287109375,
0.0279998779296875,
0.03387451171875,
0.03497314453125,
-0.061767578125,
-0.0147705078125,
-0.0225982666015625,
-0.07550048828125,
-0.019866943359375,
-0.0057830810546875,
-0.016082763671875,
-0.005992889404296875,
0.01629638671875,
-0.06463623046875,
0.0257568359375,
0.032470703125,
-0.0174560546875,
0.0450439453125,
-0.005176544189453125,
0.0204315185546875,
-0.09368896484375,
0.044525146484375,
-0.007305145263671875,
-0.00018072128295898438,
-0.06488037109375,
0.00542449951171875,
0.001811981201171875,
-0.0149383544921875,
-0.03515625,
0.053985595703125,
-0.010528564453125,
0.004467010498046875,
-0.0159454345703125,
0.00392913818359375,
0.004489898681640625,
0.049041748046875,
-0.003032684326171875,
0.05133056640625,
0.03814697265625,
-0.0447998046875,
0.0273590087890625,
0.056732177734375,
-0.022979736328125,
0.0484619140625,
-0.07452392578125,
0.0085601806640625,
-0.025146484375,
0.00809478759765625,
-0.0731201171875,
-0.03271484375,
0.0362548828125,
-0.0615234375,
0.008148193359375,
0.003826141357421875,
-0.0230865478515625,
-0.04864501953125,
-0.009796142578125,
0.01611328125,
0.048492431640625,
-0.024261474609375,
0.0279083251953125,
0.040283203125,
-0.0169219970703125,
-0.05206298828125,
-0.05047607421875,
-0.017242431640625,
-0.0224151611328125,
-0.048858642578125,
0.01149749755859375,
-0.027496337890625,
0.00591278076171875,
-0.008026123046875,
0.003307342529296875,
-0.0059356689453125,
0.00022971630096435547,
0.0285491943359375,
0.029998779296875,
-0.0119171142578125,
0.0185394287109375,
0.003688812255859375,
-0.0021343231201171875,
0.01111602783203125,
0.0225982666015625,
0.0489501953125,
-0.003299713134765625,
-0.022979736328125,
-0.04022216796875,
0.0303955078125,
0.042816162109375,
-0.0285491943359375,
0.0687255859375,
0.03326416015625,
-0.024322509765625,
0.0149993896484375,
-0.0513916015625,
-0.0057373046875,
-0.03021240234375,
0.037109375,
-0.014617919921875,
-0.04998779296875,
0.058929443359375,
0.007080078125,
-0.004119873046875,
0.04644775390625,
0.06427001953125,
0.00044465065002441406,
0.0914306640625,
0.0172271728515625,
-0.001285552978515625,
0.003833770751953125,
-0.0245208740234375,
-0.000751495361328125,
-0.0748291015625,
-0.03729248046875,
-0.034149169921875,
-0.01070404052734375,
-0.0355224609375,
-0.0384521484375,
0.036224365234375,
-0.00897979736328125,
-0.0234375,
0.017822265625,
-0.041229248046875,
0.00022125244140625,
0.030426025390625,
0.0294189453125,
0.0017147064208984375,
-0.02740478515625,
0.0015048980712890625,
0.0212860107421875,
-0.0625,
-0.0377197265625,
0.091064453125,
0.039764404296875,
0.04840087890625,
-0.0183868408203125,
0.04730224609375,
-0.0018939971923828125,
0.032257080078125,
-0.05108642578125,
0.0374755859375,
-0.0035228729248046875,
-0.049224853515625,
-0.005168914794921875,
-0.042083740234375,
-0.08343505859375,
0.01812744140625,
-0.0160369873046875,
-0.04962158203125,
0.0030307769775390625,
0.00780487060546875,
-0.03314208984375,
0.01076507568359375,
-0.0531005859375,
0.08575439453125,
-0.0134735107421875,
-0.0041656494140625,
0.0196685791015625,
-0.0662841796875,
0.03338623046875,
0.0002498626708984375,
0.0137481689453125,
-0.00689697265625,
0.007476806640625,
0.0667724609375,
-0.0211334228515625,
0.06939697265625,
-0.01175689697265625,
0.0149383544921875,
0.024139404296875,
-0.026885986328125,
0.0137176513671875,
0.00382232666015625,
-0.01070404052734375,
0.0025157928466796875,
0.039764404296875,
-0.031707763671875,
-0.046844482421875,
0.037017822265625,
-0.06781005859375,
-0.04876708984375,
-0.02752685546875,
-0.043701171875,
-0.0283660888671875,
0.00035500526428222656,
0.015167236328125,
0.032989501953125,
-0.00641632080078125,
0.006496429443359375,
0.048583984375,
-0.04541015625,
-0.0000966787338256836,
0.03076171875,
-0.0033740997314453125,
-0.03399658203125,
0.051971435546875,
0.004032135009765625,
0.020416259765625,
0.0208892822265625,
0.0252685546875,
-0.0272674560546875,
-0.03131103515625,
-0.044769287109375,
0.049591064453125,
-0.042877197265625,
-0.016326904296875,
-0.06414794921875,
-0.032501220703125,
-0.039215087890625,
0.016632080078125,
-0.007843017578125,
-0.028228759765625,
-0.026458740234375,
-0.005352020263671875,
0.0166168212890625,
0.03216552734375,
0.0252685546875,
0.01983642578125,
-0.0592041015625,
0.04351806640625,
0.01800537109375,
0.01154327392578125,
-0.0086822509765625,
-0.04156494140625,
-0.00449371337890625,
0.002895355224609375,
-0.0283660888671875,
-0.09149169921875,
0.0313720703125,
0.00916290283203125,
0.0595703125,
0.016387939453125,
0.0011625289916992188,
0.038787841796875,
-0.02813720703125,
0.07318115234375,
0.01029205322265625,
-0.04156494140625,
0.046478271484375,
-0.0033740997314453125,
0.032379150390625,
0.038848876953125,
0.047393798828125,
-0.0477294921875,
-0.03204345703125,
-0.06671142578125,
-0.04925537109375,
0.040679931640625,
0.015533447265625,
0.006378173828125,
-0.0152130126953125,
0.044647216796875,
0.007045745849609375,
0.030853271484375,
-0.045013427734375,
-0.0233306884765625,
-0.01235198974609375,
-0.037933349609375,
-0.00423431396484375,
-0.004947662353515625,
-0.019195556640625,
-0.024200439453125,
0.0540771484375,
-0.023345947265625,
0.033355712890625,
0.032135009765625,
0.0018262863159179688,
0.01067352294921875,
0.0211029052734375,
0.030029296875,
0.051422119140625,
-0.0113372802734375,
-0.00212860107421875,
0.0231475830078125,
-0.040863037109375,
-0.00724029541015625,
0.019195556640625,
-0.022491455078125,
-0.0176544189453125,
0.02813720703125,
0.07196044921875,
-0.010345458984375,
-0.07720947265625,
0.041473388671875,
-0.004268646240234375,
-0.0113372802734375,
-0.00826263427734375,
0.0152587890625,
0.02349853515625,
0.0213470458984375,
-0.0025882720947265625,
-0.01129913330078125,
0.01605224609375,
-0.050689697265625,
0.0188446044921875,
0.0197601318359375,
-0.0214080810546875,
-0.01983642578125,
0.0701904296875,
-0.0026912689208984375,
-0.01910400390625,
0.0389404296875,
0.006450653076171875,
-0.044342041015625,
0.065185546875,
0.0306854248046875,
0.035736083984375,
-0.024139404296875,
0.036163330078125,
0.0526123046875,
0.007137298583984375,
-0.0079193115234375,
0.034912109375,
0.01092529296875,
-0.039154052734375,
-0.0225830078125,
-0.0625,
-0.034027099609375,
0.02392578125,
-0.04595947265625,
0.0055084228515625,
-0.0259857177734375,
-0.0219573974609375,
0.0138397216796875,
0.00324249267578125,
-0.051055908203125,
0.03790283203125,
0.0050506591796875,
0.06256103515625,
-0.0712890625,
0.05096435546875,
0.058197021484375,
-0.0640869140625,
-0.06805419921875,
-0.0206756591796875,
-0.015350341796875,
-0.0767822265625,
0.02587890625,
0.01727294921875,
0.038055419921875,
-0.027618408203125,
-0.050323486328125,
-0.052520751953125,
0.08428955078125,
0.0275421142578125,
-0.012451171875,
-0.01433563232421875,
0.031707763671875,
0.06573486328125,
-0.03350830078125,
0.03704833984375,
0.048858642578125,
0.01299285888671875,
0.0213165283203125,
-0.0750732421875,
0.01800537109375,
-0.034271240234375,
-0.00728607177734375,
-0.0157470703125,
-0.05792236328125,
0.06500244140625,
-0.0229034423828125,
0.01015472412109375,
0.014801025390625,
0.03424072265625,
0.039581298828125,
0.033843994140625,
0.0295562744140625,
0.0489501953125,
0.0614013671875,
0.0008416175842285156,
0.09185791015625,
-0.026153564453125,
0.02850341796875,
0.05902099609375,
-0.00685882568359375,
0.05645751953125,
0.02978515625,
-0.0224151611328125,
0.0270843505859375,
0.051788330078125,
-0.01090240478515625,
0.0257568359375,
0.01922607421875,
-0.0009737014770507812,
-0.0143585205078125,
0.006725311279296875,
-0.033538818359375,
0.0302276611328125,
0.0225067138671875,
-0.041229248046875,
0.00252532958984375,
0.00897216796875,
0.01538848876953125,
-0.0174102783203125,
-0.0065460205078125,
0.054718017578125,
0.01322174072265625,
-0.07220458984375,
0.05902099609375,
-0.0128021240234375,
0.023681640625,
-0.046356201171875,
-0.00794219970703125,
-0.0262451171875,
0.0002732276916503906,
-0.014923095703125,
-0.058074951171875,
0.0153656005859375,
0.0021610260009765625,
-0.004100799560546875,
0.01513671875,
0.04229736328125,
-0.03680419921875,
-0.049224853515625,
0.00099945068359375,
0.0518798828125,
0.037384033203125,
0.00917816162109375,
-0.0753173828125,
-0.006519317626953125,
-0.00830841064453125,
-0.0244293212890625,
0.026397705078125,
0.015472412109375,
0.00879669189453125,
0.06732177734375,
0.03753662109375,
0.0143585205078125,
-0.0095672607421875,
-0.0039825439453125,
0.0771484375,
-0.03216552734375,
-0.03387451171875,
-0.03863525390625,
0.039520263671875,
-0.0119171142578125,
-0.030975341796875,
0.0447998046875,
0.043487548828125,
0.0377197265625,
-0.00852203369140625,
0.051239013671875,
-0.019287109375,
0.059295654296875,
-0.037322998046875,
0.06573486328125,
-0.054901123046875,
0.01409149169921875,
-0.0218505859375,
-0.04095458984375,
-0.01386260986328125,
0.041107177734375,
-0.0281829833984375,
0.0033168792724609375,
0.051025390625,
0.06976318359375,
0.004669189453125,
0.008270263671875,
0.0157470703125,
0.024566650390625,
0.0203857421875,
0.06048583984375,
0.033599853515625,
-0.04718017578125,
0.03192138671875,
-0.02545166015625,
-0.030914306640625,
0.004238128662109375,
-0.029754638671875,
-0.08636474609375,
-0.0472412109375,
-0.0308685302734375,
-0.041168212890625,
0.003871917724609375,
0.060272216796875,
0.041717529296875,
-0.0618896484375,
-0.00910186767578125,
0.026947021484375,
-0.00495147705078125,
-0.0214996337890625,
-0.01397705078125,
0.0511474609375,
-0.01416015625,
-0.041351318359375,
0.00817108154296875,
-0.0111236572265625,
-0.01441192626953125,
-0.00980377197265625,
0.005084991455078125,
-0.020751953125,
0.0172119140625,
0.038421630859375,
0.03021240234375,
-0.042510986328125,
-0.02850341796875,
0.0151519775390625,
-0.00870513916015625,
0.0274505615234375,
0.0272216796875,
-0.051513671875,
0.03070068359375,
0.03558349609375,
0.05987548828125,
0.044708251953125,
0.0219573974609375,
0.0194244384765625,
-0.025970458984375,
-0.0130615234375,
0.0149993896484375,
0.021942138671875,
0.03662109375,
-0.028045654296875,
0.042236328125,
0.0232391357421875,
-0.057647705078125,
-0.0482177734375,
-0.006771087646484375,
-0.098876953125,
-0.0139312744140625,
0.08856201171875,
-0.006336212158203125,
-0.037200927734375,
-0.0024623870849609375,
-0.043426513671875,
0.03350830078125,
-0.0139312744140625,
0.072265625,
0.0450439453125,
-0.04193115234375,
-0.0159912109375,
-0.06939697265625,
0.034759521484375,
0.036163330078125,
-0.09619140625,
-0.0205841064453125,
0.0168609619140625,
0.044708251953125,
0.0087127685546875,
0.06451416015625,
-0.0209808349609375,
0.0242462158203125,
-0.01080322265625,
-0.005084991455078125,
0.0004031658172607422,
0.0095977783203125,
-0.0157623291015625,
0.025909423828125,
-0.0128326416015625,
-0.0176849365234375
]
] |
dbmdz/bert-base-turkish-uncased | 2021-05-19T15:15:54.000Z | [
"transformers",
"pytorch",
"tf",
"jax",
"bert",
"tr",
"license:mit",
"endpoints_compatible",
"has_space",
"region:us"
] | null | dbmdz | null | null | dbmdz/bert-base-turkish-uncased | 14 | 7,218 | transformers | 2022-03-02T23:29:05 | ---
language: tr
license: mit
---
# 🤗 + 📚 dbmdz Turkish BERT model
In this repository the MDZ Digital Library team (dbmdz) at the Bavarian State
Library open sources an uncased model for Turkish 🎉
# 🇹🇷 BERTurk
BERTurk is a community-driven uncased BERT model for Turkish.
Some datasets used for pretraining and evaluation are contributed from the
awesome Turkish NLP community, as well as the decision for the model name: BERTurk.
## Stats
The current version of the model is trained on a filtered and sentence
segmented version of the Turkish [OSCAR corpus](https://traces1.inria.fr/oscar/),
a recent Wikipedia dump, various [OPUS corpora](http://opus.nlpl.eu/) and a
special corpus provided by [Kemal Oflazer](http://www.andrew.cmu.edu/user/ko/).
The final training corpus has a size of 35GB and 44,04,976,662 tokens.
Thanks to Google's TensorFlow Research Cloud (TFRC) we could train an uncased model
on a TPU v3-8 for 2M steps.
## Model weights
Currently only PyTorch-[Transformers](https://github.com/huggingface/transformers)
compatible weights are available. If you need access to TensorFlow checkpoints,
please raise an issue!
| Model | Downloads
| --------------------------------- | ---------------------------------------------------------------------------------------------------------------
| `dbmdz/bert-base-turkish-uncased` | [`config.json`](https://cdn.huggingface.co/dbmdz/bert-base-turkish-uncased/config.json) • [`pytorch_model.bin`](https://cdn.huggingface.co/dbmdz/bert-base-turkish-uncased/pytorch_model.bin) • [`vocab.txt`](https://cdn.huggingface.co/dbmdz/bert-base-turkish-uncased/vocab.txt)
## Usage
With Transformers >= 2.3 our BERTurk uncased model can be loaded like:
```python
from transformers import AutoModel, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("dbmdz/bert-base-turkish-uncased")
model = AutoModel.from_pretrained("dbmdz/bert-base-turkish-uncased")
```
## Results
For results on PoS tagging or NER tasks, please refer to
[this repository](https://github.com/stefan-it/turkish-bert).
# Huggingface model hub
All models are available on the [Huggingface model hub](https://huggingface.co/dbmdz).
# Contact (Bugs, Feedback, Contribution and more)
For questions about our BERT models just open an issue
[here](https://github.com/dbmdz/berts/issues/new) 🤗
# Acknowledgments
Thanks to [Kemal Oflazer](http://www.andrew.cmu.edu/user/ko/) for providing us
additional large corpora for Turkish. Many thanks to Reyyan Yeniterzi for providing
us the Turkish NER dataset for evaluation.
Research supported with Cloud TPUs from Google's TensorFlow Research Cloud (TFRC).
Thanks for providing access to the TFRC ❤️
Thanks to the generous support from the [Hugging Face](https://huggingface.co/) team,
it is possible to download both cased and uncased models from their S3 storage 🤗
| 2,880 | [
[
-0.03741455078125,
-0.05047607421875,
0.00945281982421875,
0.0196075439453125,
-0.033477783203125,
-0.018310546875,
-0.0268402099609375,
-0.031890869140625,
0.016571044921875,
0.030242919921875,
-0.045562744140625,
-0.050262451171875,
-0.050750732421875,
-0.01024627685546875,
-0.0214996337890625,
0.08807373046875,
-0.0120391845703125,
0.02215576171875,
0.004901885986328125,
-0.00572967529296875,
-0.004642486572265625,
-0.0440673828125,
-0.0289306640625,
-0.03778076171875,
0.0214080810546875,
0.005130767822265625,
0.035186767578125,
0.013946533203125,
0.035430908203125,
0.0233001708984375,
-0.01502227783203125,
-0.00957489013671875,
-0.004039764404296875,
-0.003108978271484375,
0.015777587890625,
-0.0005898475646972656,
-0.03887939453125,
-0.0078887939453125,
0.056671142578125,
0.049072265625,
-0.0241851806640625,
0.015228271484375,
0.003307342529296875,
0.052398681640625,
-0.0233612060546875,
0.013946533203125,
-0.0286102294921875,
0.007053375244140625,
-0.01348876953125,
0.01934814453125,
-0.0165252685546875,
-0.00977325439453125,
0.032073974609375,
-0.019439697265625,
0.033447265625,
-0.0233001708984375,
0.0888671875,
0.01149749755859375,
-0.02691650390625,
-0.0171661376953125,
-0.036895751953125,
0.059051513671875,
-0.072509765625,
0.0396728515625,
0.023773193359375,
0.02685546875,
-0.03656005859375,
-0.058685302734375,
-0.036895751953125,
-0.01462554931640625,
-0.004520416259765625,
0.00457763671875,
-0.022918701171875,
0.0101318359375,
0.00949859619140625,
0.036346435546875,
-0.0303192138671875,
-0.0106964111328125,
-0.0452880859375,
-0.015533447265625,
0.045379638671875,
-0.0076751708984375,
0.01373291015625,
-0.02239990234375,
-0.0261688232421875,
-0.031585693359375,
-0.03753662109375,
0.0213623046875,
0.039794921875,
0.034942626953125,
-0.0283050537109375,
0.05706787109375,
-0.007282257080078125,
0.049835205078125,
0.0165252685546875,
0.004711151123046875,
0.0305633544921875,
-0.005153656005859375,
-0.0268707275390625,
0.0037136077880859375,
0.06549072265625,
-0.00029754638671875,
-0.00008046627044677734,
-0.01407623291015625,
-0.0258636474609375,
-0.0247955322265625,
0.033935546875,
-0.06646728515625,
-0.0227813720703125,
0.030242919921875,
-0.04888916015625,
-0.0225372314453125,
0.002086639404296875,
-0.045318603515625,
-0.014495849609375,
-0.00818634033203125,
0.053863525390625,
-0.039886474609375,
-0.03521728515625,
0.020965576171875,
0.0015354156494140625,
0.03460693359375,
0.014862060546875,
-0.07928466796875,
0.0231475830078125,
0.03692626953125,
0.054107666015625,
0.01328277587890625,
-0.00682830810546875,
0.0018930435180664062,
-0.02423095703125,
-0.024261474609375,
0.05133056640625,
-0.0062713623046875,
-0.0128631591796875,
0.01555633544921875,
0.01953125,
-0.02093505859375,
-0.0217437744140625,
0.050811767578125,
-0.0325927734375,
0.040191650390625,
-0.025726318359375,
-0.047698974609375,
-0.033172607421875,
0.0013332366943359375,
-0.046112060546875,
0.09588623046875,
0.025909423828125,
-0.0706787109375,
0.031524658203125,
-0.050323486328125,
-0.0306396484375,
-0.006122589111328125,
0.0010061264038085938,
-0.06573486328125,
0.010986328125,
0.01885986328125,
0.047607421875,
-0.0009822845458984375,
0.0096893310546875,
-0.035125732421875,
-0.02203369140625,
0.009368896484375,
0.01157379150390625,
0.09075927734375,
0.0198211669921875,
-0.038055419921875,
0.01477813720703125,
-0.04608154296875,
-0.006275177001953125,
0.0140380859375,
-0.036468505859375,
0.00713348388671875,
-0.010284423828125,
0.0245819091796875,
0.0200653076171875,
0.02264404296875,
-0.055908203125,
0.0207366943359375,
-0.029144287109375,
0.03277587890625,
0.049163818359375,
-0.0220794677734375,
0.0221405029296875,
-0.0291900634765625,
0.01934814453125,
0.003391265869140625,
0.0085906982421875,
0.0165252685546875,
-0.037872314453125,
-0.07855224609375,
-0.045989990234375,
0.037078857421875,
0.0261993408203125,
-0.046295166015625,
0.05316162109375,
0.0009241104125976562,
-0.049530029296875,
-0.05487060546875,
0.0033054351806640625,
0.00787353515625,
0.042724609375,
0.026611328125,
-0.0268402099609375,
-0.059295654296875,
-0.0697021484375,
0.00658416748046875,
-0.0171661376953125,
-0.0225677490234375,
0.0219879150390625,
0.051025390625,
-0.01485443115234375,
0.052490234375,
0.00205230712890625,
-0.042083740234375,
-0.0272064208984375,
0.011627197265625,
0.04296875,
0.04681396484375,
0.05694580078125,
-0.027191162109375,
-0.027313232421875,
-0.0193023681640625,
-0.0556640625,
0.01446533203125,
0.005840301513671875,
-0.01355743408203125,
0.054931640625,
0.019561767578125,
-0.06573486328125,
0.0287628173828125,
0.036285400390625,
-0.044464111328125,
0.052490234375,
-0.0268096923828125,
-0.0006060600280761719,
-0.08697509765625,
0.025482177734375,
0.001964569091796875,
-0.0100860595703125,
-0.037567138671875,
0.0055084228515625,
-0.001850128173828125,
0.004558563232421875,
-0.038543701171875,
0.042388916015625,
-0.0233917236328125,
-0.010284423828125,
-0.0023937225341796875,
-0.030487060546875,
-0.008880615234375,
0.049774169921875,
0.0215606689453125,
0.04718017578125,
0.04669189453125,
-0.03668212890625,
0.0255584716796875,
0.036529541015625,
-0.059844970703125,
0.007537841796875,
-0.06634521484375,
0.0034027099609375,
0.004047393798828125,
0.021453857421875,
-0.05914306640625,
-0.01139068603515625,
0.033935546875,
-0.04583740234375,
0.044677734375,
-0.0328369140625,
-0.0604248046875,
-0.030975341796875,
-0.0168609619140625,
-0.002452850341796875,
0.05487060546875,
-0.04998779296875,
0.053375244140625,
0.024444580078125,
-0.0094146728515625,
-0.056427001953125,
-0.054351806640625,
-0.007747650146484375,
-0.0283966064453125,
-0.062744140625,
0.03271484375,
-0.00852203369140625,
0.0025424957275390625,
0.004947662353515625,
-0.0047454833984375,
-0.0159759521484375,
-0.0019044876098632812,
0.01302337646484375,
0.03387451171875,
-0.01727294921875,
0.0027008056640625,
0.0005536079406738281,
0.0097198486328125,
0.00665283203125,
-0.0222320556640625,
0.039337158203125,
-0.038543701171875,
-0.00423431396484375,
-0.032867431640625,
0.01922607421875,
0.033111572265625,
-0.0033702850341796875,
0.089599609375,
0.07574462890625,
-0.03564453125,
0.01215362548828125,
-0.054840087890625,
-0.0187225341796875,
-0.03466796875,
0.01438140869140625,
-0.026519775390625,
-0.07012939453125,
0.0552978515625,
0.0234375,
0.0276031494140625,
0.04901123046875,
0.0625,
-0.03466796875,
0.0733642578125,
0.07086181640625,
-0.0003581047058105469,
0.042724609375,
-0.0298614501953125,
0.004055023193359375,
-0.056732177734375,
-0.0285491943359375,
-0.036895751953125,
-0.00826263427734375,
-0.049163818359375,
-0.01116943359375,
0.01806640625,
0.00988006591796875,
-0.031951904296875,
0.033966064453125,
-0.040863037109375,
-0.0011816024780273438,
0.048858642578125,
0.0198211669921875,
-0.01611328125,
0.03131103515625,
-0.0322265625,
0.0028018951416015625,
-0.05609130859375,
-0.0328369140625,
0.09649658203125,
0.040252685546875,
0.0355224609375,
0.01473236083984375,
0.055877685546875,
0.0185394287109375,
0.00923919677734375,
-0.043243408203125,
0.023468017578125,
-0.00980377197265625,
-0.072021484375,
-0.01239776611328125,
-0.0233612060546875,
-0.076171875,
0.0159149169921875,
-0.03094482421875,
-0.06756591796875,
0.0163726806640625,
-0.0024566650390625,
-0.031707763671875,
0.043212890625,
-0.049041748046875,
0.0740966796875,
0.0020008087158203125,
-0.0132293701171875,
-0.01129150390625,
-0.05267333984375,
0.0164794921875,
0.0083770751953125,
-0.01334381103515625,
-0.0041046142578125,
0.021331787109375,
0.0760498046875,
-0.052398681640625,
0.05120849609375,
-0.0229339599609375,
0.0009207725524902344,
0.028594970703125,
-0.00699615478515625,
0.02459716796875,
-0.01544189453125,
-0.00879669189453125,
0.038604736328125,
0.0210418701171875,
-0.05328369140625,
-0.0188751220703125,
0.0382080078125,
-0.08502197265625,
-0.02825927734375,
-0.049530029296875,
-0.033050537109375,
0.0005474090576171875,
0.02166748046875,
0.0206756591796875,
0.0204010009765625,
-0.01052093505859375,
0.0238494873046875,
0.061767578125,
-0.030487060546875,
0.037750244140625,
0.044342041015625,
-0.0007319450378417969,
-0.02215576171875,
0.040435791015625,
-0.005664825439453125,
-0.01739501953125,
0.00919342041015625,
0.0012826919555664062,
-0.03704833984375,
-0.03826904296875,
-0.04034423828125,
0.03314208984375,
-0.04290771484375,
-0.006927490234375,
-0.058624267578125,
-0.02703857421875,
-0.056427001953125,
0.008636474609375,
-0.0423583984375,
-0.04248046875,
-0.02056884765625,
-0.00675201416015625,
0.04095458984375,
0.047943115234375,
-0.026031494140625,
0.017791748046875,
-0.046356201171875,
0.005443572998046875,
0.0105743408203125,
0.039215087890625,
-0.003055572509765625,
-0.060882568359375,
-0.02447509765625,
0.0027637481689453125,
-0.01264190673828125,
-0.05072021484375,
0.043853759765625,
0.004669189453125,
0.042083740234375,
0.034027099609375,
0.00861358642578125,
0.034271240234375,
-0.0335693359375,
0.044677734375,
0.00444793701171875,
-0.048248291015625,
0.0217132568359375,
-0.03521728515625,
0.00917816162109375,
0.0361328125,
0.028350830078125,
-0.033233642578125,
-0.006664276123046875,
-0.058258056640625,
-0.0660400390625,
0.0677490234375,
0.0338134765625,
0.0136260986328125,
0.01554107666015625,
0.0211334228515625,
0.006256103515625,
0.0185394287109375,
-0.05462646484375,
-0.019500732421875,
-0.03570556640625,
-0.0155792236328125,
-0.004058837890625,
-0.036712646484375,
-0.0142059326171875,
-0.0386962890625,
0.07696533203125,
0.0203399658203125,
0.0589599609375,
0.0207366943359375,
-0.01482391357421875,
-0.018310546875,
0.0088653564453125,
0.04632568359375,
0.031402587890625,
-0.059112548828125,
-0.013580322265625,
0.00859832763671875,
-0.03778076171875,
-0.0182037353515625,
0.04461669921875,
-0.004016876220703125,
0.01461029052734375,
0.0189056396484375,
0.0625,
0.0028514862060546875,
-0.035919189453125,
0.03271484375,
-0.023406982421875,
-0.032867431640625,
-0.055908203125,
-0.0209197998046875,
0.005146026611328125,
0.0310821533203125,
0.036895751953125,
-0.0134735107421875,
-0.0036754608154296875,
-0.0167236328125,
0.0211944580078125,
0.0335693359375,
-0.034027099609375,
-0.0181884765625,
0.036895751953125,
0.005626678466796875,
0.004558563232421875,
0.078125,
0.00254058837890625,
-0.045562744140625,
0.045562744140625,
0.02069091796875,
0.0657958984375,
-0.00814056396484375,
0.0171051025390625,
0.042877197265625,
0.034423828125,
0.005512237548828125,
0.0174713134765625,
-0.01511383056640625,
-0.058349609375,
-0.023834228515625,
-0.06610107421875,
-0.01294708251953125,
0.033538818359375,
-0.05316162109375,
0.0215911865234375,
-0.03656005859375,
-0.0174560546875,
-0.001689910888671875,
0.042236328125,
-0.052276611328125,
0.00852203369140625,
0.01861572265625,
0.07440185546875,
-0.06634521484375,
0.073974609375,
0.06768798828125,
-0.038116455078125,
-0.051849365234375,
-0.036041259765625,
-0.0151519775390625,
-0.05682373046875,
0.03021240234375,
0.01401519775390625,
0.0221405029296875,
-0.01044464111328125,
-0.040313720703125,
-0.062408447265625,
0.07391357421875,
0.0206756591796875,
-0.00414276123046875,
0.0030956268310546875,
0.0082855224609375,
0.0433349609375,
-0.015716552734375,
0.03033447265625,
0.037872314453125,
0.0223541259765625,
0.0168304443359375,
-0.061553955078125,
0.005767822265625,
-0.0386962890625,
-0.0142364501953125,
0.0059051513671875,
-0.042144775390625,
0.06756591796875,
-0.0130157470703125,
-0.00603485107421875,
0.0200042724609375,
0.062408447265625,
0.0276031494140625,
-0.00447845458984375,
0.03460693359375,
0.060302734375,
0.034759521484375,
-0.005474090576171875,
0.07763671875,
-0.02056884765625,
0.039337158203125,
0.055572509765625,
0.015960693359375,
0.04876708984375,
0.0303802490234375,
-0.0211944580078125,
0.055572509765625,
0.08502197265625,
-0.02130126953125,
0.0384521484375,
-0.0038204193115234375,
-0.0258941650390625,
-0.02783203125,
-0.00016891956329345703,
-0.04095458984375,
0.02691650390625,
0.0178680419921875,
-0.024658203125,
-0.0256195068359375,
-0.0120849609375,
0.015869140625,
-0.034576416015625,
-0.00730133056640625,
0.04296875,
0.004047393798828125,
-0.0286102294921875,
0.053924560546875,
0.0182647705078125,
0.048431396484375,
-0.05291748046875,
0.003612518310546875,
-0.0172882080078125,
0.018829345703125,
-0.00827789306640625,
-0.040252685546875,
0.021087646484375,
-0.0006771087646484375,
-0.0086212158203125,
-0.0196990966796875,
0.05804443359375,
-0.0345458984375,
-0.052703857421875,
0.01244354248046875,
0.022918701171875,
0.03289794921875,
-0.004741668701171875,
-0.08270263671875,
-0.005924224853515625,
-0.005573272705078125,
-0.041046142578125,
0.035247802734375,
0.015960693359375,
0.007965087890625,
0.055572509765625,
0.04833984375,
-0.0068359375,
0.00916290283203125,
-0.0007610321044921875,
0.06524658203125,
-0.03375244140625,
-0.0284271240234375,
-0.0379638671875,
0.047332763671875,
0.0109405517578125,
-0.01165008544921875,
0.05078125,
0.03564453125,
0.0657958984375,
-0.0120697021484375,
0.045989990234375,
-0.0248260498046875,
0.0341796875,
-0.0232086181640625,
0.083984375,
-0.052642822265625,
-0.01348876953125,
-0.0272369384765625,
-0.057403564453125,
-0.0135955810546875,
0.072021484375,
-0.0145263671875,
0.0190887451171875,
0.032684326171875,
0.045501708984375,
0.001308441162109375,
-0.01349639892578125,
0.0065460205078125,
0.022003173828125,
0.005069732666015625,
0.03973388671875,
0.03271484375,
-0.048583984375,
0.041473388671875,
-0.0465087890625,
-0.00998687744140625,
-0.0192413330078125,
-0.0604248046875,
-0.095458984375,
-0.057220458984375,
-0.022796630859375,
-0.049468994140625,
-0.001201629638671875,
0.07855224609375,
0.0615234375,
-0.0772705078125,
-0.029327392578125,
-0.0033168792724609375,
0.0021839141845703125,
-0.0154266357421875,
-0.01528167724609375,
0.05712890625,
-0.022552490234375,
-0.05291748046875,
0.01323699951171875,
-0.0216522216796875,
0.0286407470703125,
-0.0124359130859375,
-0.007686614990234375,
-0.0273895263671875,
0.0023097991943359375,
0.02734375,
0.035186767578125,
-0.040924072265625,
-0.0097198486328125,
0.0010576248168945312,
-0.00614166259765625,
0.004009246826171875,
0.038787841796875,
-0.0477294921875,
0.0343017578125,
0.030853271484375,
0.0295257568359375,
0.060516357421875,
-0.0252227783203125,
0.039337158203125,
-0.047393798828125,
0.03094482421875,
0.00836181640625,
0.0426025390625,
0.034027099609375,
-0.01432037353515625,
0.031036376953125,
0.004241943359375,
-0.03546142578125,
-0.06451416015625,
-0.011627197265625,
-0.07763671875,
-0.022796630859375,
0.06683349609375,
-0.0247650146484375,
-0.027557373046875,
0.004085540771484375,
0.0029659271240234375,
0.049652099609375,
-0.034393310546875,
0.088623046875,
0.06524658203125,
-0.0010709762573242188,
-0.0234527587890625,
-0.034423828125,
0.045623779296875,
0.04925537109375,
-0.028106689453125,
-0.01439666748046875,
0.019775390625,
0.035369873046875,
-0.011199951171875,
0.03448486328125,
-0.0208740234375,
0.0186309814453125,
-0.00994110107421875,
0.048431396484375,
-0.0150146484375,
-0.0095062255859375,
-0.0312042236328125,
-0.017303466796875,
-0.0155487060546875,
-0.0167083740234375
]
] |
Delcos/Mistral-Pygmalion-7b | 2023-10-19T17:43:10.000Z | [
"transformers",
"pytorch",
"llama",
"text-generation",
"Mistral",
"Pygmalion",
"llama-2",
"llama-2-7b",
"en",
"license:cc-by-nc-nd-4.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | Delcos | null | null | Delcos/Mistral-Pygmalion-7b | 7 | 7,218 | transformers | 2023-10-09T21:36:41 | ---
license: cc-by-nc-nd-4.0
language:
- en
library_name: transformers
pipeline_tag: text-generation
tags:
- Mistral
- Pygmalion
- llama-2
- llama-2-7b
---
# MistralPy-7b
This is a merger focusing on preserving the roleplay abilities of Pygmalion while gaining the improved results from Mistral. This model works best for roleplay but is still fairly capable assistant. The smaller (7b) size does mean it isn't perfect at more complex reasoning tasks, but this should be addressed in the larger version that I'll upload soon.
[GGUF version done by TheBloke](https://huggingface.co/TheBloke/Mistral-Pygmalion-7B-GGUF)
### Prompt Template
```
### Instruction:
{Prompt & Backstory}
### Assistant:
{Output}
```
Example:
```
### Instruction:
You are Sally, a fun 19 year old woman. Her favorite animal is "cat". Her favoritate color is "blue". She enjoys grape juice and cake.
### Assistant:
Sally: Hi, how are you?
User: Okay, you?
```
# Send a message
[Steam](https://steamcommunity.com/id/delcos/)
#### Discord: delcos69 | 1,025 | [
[
-0.017852783203125,
-0.045989990234375,
0.0178680419921875,
0.0310211181640625,
-0.032318115234375,
-0.034576416015625,
-0.0048980712890625,
-0.045440673828125,
0.0216522216796875,
0.0325927734375,
-0.027679443359375,
0.0027713775634765625,
-0.0543212890625,
0.004955291748046875,
-0.023101806640625,
0.05706787109375,
0.012664794921875,
0.004428863525390625,
-0.010498046875,
0.01377105712890625,
-0.0531005859375,
-0.041107177734375,
-0.0997314453125,
-0.050262451171875,
0.0390625,
0.0156707763671875,
0.059814453125,
0.04437255859375,
0.011505126953125,
0.039520263671875,
-0.03094482421875,
0.0176544189453125,
-0.0391845703125,
0.0284881591796875,
-0.02276611328125,
-0.03680419921875,
-0.04937744140625,
0.001407623291015625,
0.02215576171875,
0.029327392578125,
-0.03955078125,
0.00943756103515625,
0.0162506103515625,
0.0178985595703125,
-0.03656005859375,
0.03424072265625,
-0.0196075439453125,
-0.0006074905395507812,
0.015899658203125,
0.0140380859375,
-0.0031299591064453125,
-0.0232696533203125,
0.0218963623046875,
-0.0782470703125,
0.01837158203125,
-0.00818634033203125,
0.060577392578125,
0.0198822021484375,
-0.04144287109375,
-0.006732940673828125,
-0.0450439453125,
0.041229248046875,
-0.04583740234375,
0.01038360595703125,
0.03997802734375,
0.04095458984375,
-0.03424072265625,
-0.08306884765625,
-0.0235595703125,
-0.020660400390625,
0.01103973388671875,
0.005096435546875,
-0.043731689453125,
-0.0001208186149597168,
0.0249176025390625,
0.03955078125,
-0.027587890625,
-0.0177154541015625,
-0.0565185546875,
-0.0261077880859375,
0.0301971435546875,
0.008148193359375,
0.03857421875,
0.00547027587890625,
-0.031524658203125,
0.00765228271484375,
-0.011138916015625,
0.0102691650390625,
0.0160980224609375,
-0.0132293701171875,
-0.012115478515625,
0.043182373046875,
-0.0053558349609375,
0.0537109375,
0.04425048828125,
-0.0033321380615234375,
0.0001786947250366211,
-0.0255889892578125,
-0.03607177734375,
-0.0023517608642578125,
0.046539306640625,
0.048492431640625,
0.0105133056640625,
-0.00554656982421875,
0.0014476776123046875,
0.00698089599609375,
0.0350341796875,
-0.053802490234375,
-0.059234619140625,
0.0198211669921875,
-0.050750732421875,
-0.017303466796875,
0.01502227783203125,
-0.01100921630859375,
-0.044586181640625,
-0.02679443359375,
0.034210205078125,
-0.04534912109375,
-0.04010009765625,
0.0175018310546875,
-0.058746337890625,
-0.01369476318359375,
0.07757568359375,
-0.054046630859375,
0.017547607421875,
0.035369873046875,
0.0452880859375,
0.005031585693359375,
-0.047576904296875,
-0.0310821533203125,
0.01128387451171875,
-0.036285400390625,
0.035888671875,
-0.00276947021484375,
-0.0545654296875,
-0.03570556640625,
0.015655517578125,
0.001644134521484375,
-0.06256103515625,
0.030609130859375,
-0.01207733154296875,
0.045684814453125,
-0.01018524169921875,
-0.00991058349609375,
-0.0166168212890625,
0.014892578125,
-0.06866455078125,
0.0621337890625,
0.004016876220703125,
-0.049835205078125,
0.0214996337890625,
-0.041259765625,
-0.00878143310546875,
0.011566162109375,
0.0159759521484375,
0.00489044189453125,
0.0095062255859375,
0.00016260147094726562,
0.02276611328125,
-0.018218994140625,
0.00970458984375,
-0.0338134765625,
-0.0352783203125,
0.032379150390625,
-0.0288543701171875,
0.0616455078125,
0.0202178955078125,
-0.0084991455078125,
0.003910064697265625,
-0.041412353515625,
-0.00920867919921875,
0.0057373046875,
-0.0297088623046875,
-0.01485443115234375,
-0.0286102294921875,
0.00817108154296875,
0.023590087890625,
0.0660400390625,
-0.004291534423828125,
0.04296875,
0.0018396377563476562,
0.0223846435546875,
0.049407958984375,
0.005157470703125,
0.057586669921875,
-0.04852294921875,
0.033050537109375,
-0.00936126708984375,
0.05743408203125,
0.036834716796875,
-0.05609130859375,
-0.0458984375,
-0.045318603515625,
0.00787353515625,
0.0297088623046875,
-0.07452392578125,
0.052154541015625,
0.031951904296875,
-0.047332763671875,
-0.03240966796875,
-0.0021820068359375,
0.0222015380859375,
0.0196533203125,
0.0179290771484375,
-0.030181884765625,
-0.044921875,
-0.059356689453125,
-0.00013256072998046875,
-0.042510986328125,
0.01168060302734375,
0.0248870849609375,
-0.0112457275390625,
-0.040618896484375,
0.037994384765625,
-0.0286407470703125,
-0.021881103515625,
-0.035797119140625,
-0.0004649162292480469,
0.038604736328125,
0.05096435546875,
0.041748046875,
-0.056610107421875,
-0.0146636962890625,
0.0009541511535644531,
-0.058929443359375,
-0.0250244140625,
0.0091705322265625,
-0.026397705078125,
0.0024623870849609375,
0.007720947265625,
-0.0726318359375,
0.0390625,
0.041412353515625,
-0.056671142578125,
0.0272216796875,
-0.04754638671875,
0.045806884765625,
-0.0992431640625,
0.01031494140625,
-0.0133819580078125,
-0.013671875,
-0.03594970703125,
0.007717132568359375,
-0.022369384765625,
0.0016231536865234375,
-0.04217529296875,
0.045379638671875,
-0.061004638671875,
0.01212310791015625,
-0.034759521484375,
-0.019561767578125,
-0.005619049072265625,
0.003002166748046875,
-0.0113525390625,
0.039642333984375,
0.062744140625,
-0.052764892578125,
0.0626220703125,
0.0291900634765625,
0.0217742919921875,
0.036102294921875,
-0.056610107421875,
0.00803375244140625,
0.010223388671875,
0.03607177734375,
-0.06732177734375,
-0.0149078369140625,
0.045989990234375,
-0.045440673828125,
0.0250396728515625,
-0.048492431640625,
-0.019256591796875,
-0.02349853515625,
-0.02923583984375,
0.034088134765625,
0.046112060546875,
-0.051055908203125,
0.055999755859375,
0.0216522216796875,
-0.01580810546875,
-0.040771484375,
-0.047943115234375,
0.0155792236328125,
-0.02587890625,
-0.05511474609375,
0.0235595703125,
-0.012939453125,
-0.0087432861328125,
-0.0005507469177246094,
-0.0138397216796875,
-0.029327392578125,
-0.0045928955078125,
0.040679931640625,
0.029205322265625,
0.001415252685546875,
-0.03021240234375,
-0.00379180908203125,
-0.0158538818359375,
-0.00949859619140625,
0.002826690673828125,
0.03887939453125,
0.00914764404296875,
0.0146636962890625,
-0.041473388671875,
0.0116729736328125,
0.07366943359375,
0.0002872943878173828,
0.04119873046875,
0.06494140625,
-0.025909423828125,
-0.00911712646484375,
-0.034576416015625,
-0.0008754730224609375,
-0.036773681640625,
0.004207611083984375,
-0.0269775390625,
-0.05999755859375,
0.051849365234375,
0.00628662109375,
-0.0155181884765625,
0.043243408203125,
0.03912353515625,
0.01202392578125,
0.06298828125,
0.029998779296875,
-0.002109527587890625,
0.048736572265625,
-0.0283355712890625,
-0.0013484954833984375,
-0.05865478515625,
-0.0316162109375,
-0.022003173828125,
-0.00789642333984375,
-0.027679443359375,
-0.03521728515625,
0.022125244140625,
0.049652099609375,
-0.008819580078125,
0.047882080078125,
-0.0498046875,
0.0109100341796875,
0.045166015625,
0.0248870849609375,
0.006359100341796875,
-0.0214996337890625,
0.0223846435546875,
0.00782012939453125,
-0.060882568359375,
-0.022918701171875,
0.03662109375,
0.037445068359375,
0.07647705078125,
0.05035400390625,
0.07427978515625,
0.005008697509765625,
-0.005084991455078125,
-0.03912353515625,
0.04638671875,
-0.0036907196044921875,
-0.048095703125,
-0.0251922607421875,
-0.0362548828125,
-0.08251953125,
0.03277587890625,
-0.03125,
-0.06488037109375,
0.0119171142578125,
0.02874755859375,
-0.00836181640625,
-0.00443267822265625,
-0.09222412109375,
0.0767822265625,
0.0036468505859375,
-0.0267333984375,
-0.017578125,
-0.0255889892578125,
0.04241943359375,
0.0279541015625,
-0.01448822021484375,
0.020965576171875,
-0.00848388671875,
0.03814697265625,
-0.041412353515625,
0.054656982421875,
-0.01617431640625,
-0.0078125,
0.021514892578125,
0.0295562744140625,
0.020599365234375,
0.0227203369140625,
0.024078369140625,
-0.0003647804260253906,
0.0272979736328125,
-0.018035888671875,
-0.04925537109375,
0.0733642578125,
-0.06939697265625,
-0.045135498046875,
-0.0484619140625,
-0.0321044921875,
0.023590087890625,
0.005855560302734375,
0.030609130859375,
0.027618408203125,
-0.016357421875,
0.01036834716796875,
0.0289154052734375,
-0.00254058837890625,
0.0309906005859375,
0.0223846435546875,
-0.0380859375,
-0.04144287109375,
0.05718994140625,
0.004329681396484375,
-0.000438690185546875,
-0.012176513671875,
0.0240478515625,
-0.04010009765625,
-0.01056671142578125,
-0.0258331298828125,
0.01294708251953125,
-0.031341552734375,
-0.00853729248046875,
-0.0335693359375,
-0.007007598876953125,
-0.03790283203125,
0.0017118453979492188,
-0.0003311634063720703,
-0.03875732421875,
-0.0450439453125,
0.0264129638671875,
0.0234527587890625,
0.07110595703125,
-0.0259857177734375,
0.016204833984375,
-0.05859375,
0.0435791015625,
0.044830322265625,
0.0023708343505859375,
0.006134033203125,
-0.03973388671875,
-0.0018291473388671875,
-0.005077362060546875,
-0.01132965087890625,
-0.07330322265625,
0.019989013671875,
-0.00678253173828125,
0.05877685546875,
0.0270843505859375,
-0.007259368896484375,
0.05145263671875,
-0.0224151611328125,
0.0819091796875,
0.0386962890625,
-0.05780029296875,
0.01422882080078125,
-0.03375244140625,
0.016510009765625,
0.0151824951171875,
0.00214385986328125,
-0.0288848876953125,
-0.0228424072265625,
-0.08612060546875,
-0.043548583984375,
0.05657958984375,
0.03814697265625,
-0.0117950439453125,
0.0159912109375,
0.016754150390625,
-0.01153564453125,
0.01678466796875,
-0.03594970703125,
-0.022125244140625,
-0.0117645263671875,
0.0099639892578125,
0.0034809112548828125,
-0.005893707275390625,
-0.01538848876953125,
-0.040924072265625,
0.058380126953125,
-0.0022029876708984375,
0.03363037109375,
0.0019464492797851562,
0.0137176513671875,
-0.00830841064453125,
-0.0085601806640625,
0.032440185546875,
0.05950927734375,
-0.0158538818359375,
0.0225677490234375,
0.0158538818359375,
-0.0521240234375,
0.00031948089599609375,
0.032806396484375,
0.032867431640625,
-0.0019683837890625,
0.03399658203125,
0.034698486328125,
0.0007739067077636719,
-0.00931549072265625,
0.04168701171875,
-0.02783203125,
-0.002140045166015625,
-0.01142120361328125,
0.05029296875,
0.0156097412109375,
0.05816650390625,
0.020050048828125,
0.01403045654296875,
0.0352783203125,
-0.05029296875,
0.01398468017578125,
0.016021728515625,
-0.011505126953125,
-0.011566162109375,
0.050262451171875,
0.007747650146484375,
-0.027862548828125,
0.0207061767578125,
0.00848388671875,
-0.038177490234375,
0.059051513671875,
0.058563232421875,
0.059844970703125,
-0.033538818359375,
0.0126495361328125,
0.0059356689453125,
0.01291656494140625,
-0.033966064453125,
0.03118896484375,
-0.005413055419921875,
-0.0377197265625,
-0.005962371826171875,
-0.005924224853515625,
-0.0224609375,
0.0019445419311523438,
-0.034149169921875,
0.021942138671875,
-0.0377197265625,
-0.04132080078125,
-0.0078125,
-0.00023949146270751953,
-0.024810791015625,
0.0247344970703125,
-0.009979248046875,
0.05938720703125,
-0.0504150390625,
0.06396484375,
0.08538818359375,
-0.04595947265625,
-0.0667724609375,
-0.01421356201171875,
-0.005718231201171875,
-0.03662109375,
0.02764892578125,
-0.0146942138671875,
0.0035457611083984375,
-0.011077880859375,
-0.0556640625,
-0.046295166015625,
0.11175537109375,
0.031219482421875,
-0.012054443359375,
-0.01763916015625,
-0.051910400390625,
0.03570556640625,
-0.0235748291015625,
0.0293426513671875,
0.018890380859375,
0.02099609375,
0.0289306640625,
-0.09857177734375,
-0.0019855499267578125,
-0.0225067138671875,
0.0191192626953125,
-0.005908966064453125,
-0.05242919921875,
0.0823974609375,
0.00897979736328125,
-0.0037689208984375,
0.02911376953125,
0.046875,
0.03143310546875,
0.013275146484375,
0.06060791015625,
0.046905517578125,
0.032440185546875,
-0.00957489013671875,
0.06884765625,
-0.01666259765625,
0.040130615234375,
0.06451416015625,
0.0041961669921875,
0.0278167724609375,
0.038909912109375,
-0.0008873939514160156,
0.00948333740234375,
0.054931640625,
0.01666259765625,
0.0277862548828125,
0.0089111328125,
-0.005702972412109375,
-0.0168609619140625,
-0.012664794921875,
-0.03857421875,
0.00087738037109375,
-0.0019388198852539062,
-0.025177001953125,
-0.01415252685546875,
-0.0153045654296875,
-0.00426483154296875,
-0.0162353515625,
-0.0268402099609375,
0.060150146484375,
0.01322174072265625,
-0.04962158203125,
0.041900634765625,
-0.005817413330078125,
0.05474853515625,
-0.036468505859375,
-0.041748046875,
-0.019012451171875,
0.0248260498046875,
-0.0234527587890625,
-0.06329345703125,
-0.016876220703125,
-0.034210205078125,
-0.0157470703125,
0.00545501708984375,
0.07073974609375,
-0.061492919921875,
0.0009250640869140625,
0.0184326171875,
0.039276123046875,
0.01641845703125,
-0.006381988525390625,
-0.07257080078125,
0.0161895751953125,
-0.0035495758056640625,
0.026153564453125,
0.00899505615234375,
0.048828125,
-0.00853729248046875,
0.048736572265625,
0.0367431640625,
0.01297760009765625,
0.00872039794921875,
0.0028247833251953125,
0.060302734375,
-0.02252197265625,
-0.05328369140625,
-0.039276123046875,
0.04388427734375,
-0.00431060791015625,
-0.049041748046875,
0.055755615234375,
0.044921875,
0.0308837890625,
-0.0311431884765625,
0.037506103515625,
-0.00884246826171875,
0.006076812744140625,
-0.03375244140625,
0.05645751953125,
-0.044036865234375,
-0.01165008544921875,
-0.0253753662109375,
-0.0819091796875,
0.003154754638671875,
0.07061767578125,
0.01611328125,
0.0159759521484375,
0.03662109375,
0.055938720703125,
-0.035400390625,
-0.015838623046875,
0.03912353515625,
0.006969451904296875,
0.0196685791015625,
0.0570068359375,
0.1021728515625,
-0.042633056640625,
0.0180511474609375,
-0.0020198822021484375,
-0.033905029296875,
-0.024078369140625,
-0.04559326171875,
-0.08251953125,
-0.04656982421875,
-0.0228424072265625,
-0.07501220703125,
-0.00032401084899902344,
0.076171875,
0.04852294921875,
-0.0141448974609375,
-0.03076171875,
0.01117706298828125,
0.004863739013671875,
-0.020111083984375,
-0.01459503173828125,
0.0028209686279296875,
0.01165008544921875,
-0.056427001953125,
0.0167694091796875,
0.00405120849609375,
0.030609130859375,
-0.0175323486328125,
-0.01448822021484375,
-0.01343536376953125,
-0.0022563934326171875,
0.032501220703125,
0.0181427001953125,
-0.0570068359375,
-0.0217437744140625,
-0.0109710693359375,
-0.0286865234375,
-0.0309600830078125,
0.045440673828125,
-0.02618408203125,
0.0291290283203125,
0.049896240234375,
0.0213165283203125,
0.044158935546875,
0.0112457275390625,
0.056427001953125,
-0.02728271484375,
0.0192718505859375,
-0.0100860595703125,
0.040283203125,
0.0150604248046875,
-0.045135498046875,
0.045745849609375,
0.03643798828125,
-0.041290283203125,
-0.03857421875,
-0.00872039794921875,
-0.09967041015625,
-0.0246429443359375,
0.10626220703125,
0.0092315673828125,
-0.0408935546875,
0.0017595291137695312,
-0.0877685546875,
0.03607177734375,
-0.061370849609375,
0.03680419921875,
0.039398193359375,
-0.0186004638671875,
-0.0207672119140625,
-0.0008378028869628906,
0.044342041015625,
-0.002368927001953125,
-0.046295166015625,
-0.014892578125,
0.06512451171875,
0.0254364013671875,
0.020751953125,
0.056915283203125,
0.00756072998046875,
0.06402587890625,
-0.006252288818359375,
0.01849365234375,
-0.0287933349609375,
-0.035888671875,
-0.039764404296875,
-0.0005078315734863281,
0.0281219482421875,
-0.042572021484375
]
] |
flair/upos-multi | 2023-03-06T11:18:52.000Z | [
"flair",
"pytorch",
"token-classification",
"sequence-tagger-model",
"en",
"de",
"fr",
"it",
"nl",
"pl",
"es",
"sv",
"da",
"no",
"fi",
"cs",
"dataset:ontonotes",
"region:us"
] | token-classification | flair | null | null | flair/upos-multi | 6 | 7,214 | flair | 2022-03-02T23:29:05 | ---
tags:
- flair
- token-classification
- sequence-tagger-model
language:
- en
- de
- fr
- it
- nl
- pl
- es
- sv
- da
- no
- fi
- cs
datasets:
- ontonotes
widget:
- text: "Ich liebe Berlin, as they say"
---
## Multilingual Universal Part-of-Speech Tagging in Flair (default model)
This is the default multilingual universal part-of-speech tagging model that ships with [Flair](https://github.com/flairNLP/flair/).
F1-Score: **98,47** (12 UD Treebanks covering English, German, French, Italian, Dutch, Polish, Spanish, Swedish, Danish, Norwegian, Finnish and Czech)
Predicts universal POS tags:
| **tag** | **meaning** |
|---------------------------------|-----------|
|ADJ | adjective |
| ADP | adposition |
| ADV | adverb |
| AUX | auxiliary |
| CCONJ | coordinating conjunction |
| DET | determiner |
| INTJ | interjection |
| NOUN | noun |
| NUM | numeral |
| PART | particle |
| PRON | pronoun |
| PROPN | proper noun |
| PUNCT | punctuation |
| SCONJ | subordinating conjunction |
| SYM | symbol |
| VERB | verb |
| X | other |
Based on [Flair embeddings](https://www.aclweb.org/anthology/C18-1139/) and LSTM-CRF.
---
### Demo: How to use in Flair
Requires: **[Flair](https://github.com/flairNLP/flair/)** (`pip install flair`)
```python
from flair.data import Sentence
from flair.models import SequenceTagger
# load tagger
tagger = SequenceTagger.load("flair/upos-multi")
# make example sentence
sentence = Sentence("Ich liebe Berlin, as they say. ")
# predict POS tags
tagger.predict(sentence)
# print sentence
print(sentence)
# iterate over tokens and print the predicted POS label
print("The following POS tags are found:")
for token in sentence:
print(token.get_label("upos"))
```
This yields the following output:
```
Token[0]: "Ich" → PRON (0.9999)
Token[1]: "liebe" → VERB (0.9999)
Token[2]: "Berlin" → PROPN (0.9997)
Token[3]: "," → PUNCT (1.0)
Token[4]: "as" → SCONJ (0.9991)
Token[5]: "they" → PRON (0.9998)
Token[6]: "say" → VERB (0.9998)
Token[7]: "." → PUNCT (1.0)
```
So, the words "*Ich*" and "*they*" are labeled as **pronouns** (PRON), while "*liebe*" and "*say*" are labeled as **verbs** (VERB) in the multilingual sentence "*Ich liebe Berlin, as they say*".
---
### Training: Script to train this model
The following Flair script was used to train this model:
```python
from flair.data import MultiCorpus
from flair.datasets import UD_ENGLISH, UD_GERMAN, UD_FRENCH, UD_ITALIAN, UD_POLISH, UD_DUTCH, UD_CZECH, \
UD_DANISH, UD_SPANISH, UD_SWEDISH, UD_NORWEGIAN, UD_FINNISH
from flair.embeddings import StackedEmbeddings, FlairEmbeddings
# 1. make a multi corpus consisting of 12 UD treebanks (in_memory=False here because this corpus becomes large)
corpus = MultiCorpus([
UD_ENGLISH(in_memory=False),
UD_GERMAN(in_memory=False),
UD_DUTCH(in_memory=False),
UD_FRENCH(in_memory=False),
UD_ITALIAN(in_memory=False),
UD_SPANISH(in_memory=False),
UD_POLISH(in_memory=False),
UD_CZECH(in_memory=False),
UD_DANISH(in_memory=False),
UD_SWEDISH(in_memory=False),
UD_NORWEGIAN(in_memory=False),
UD_FINNISH(in_memory=False),
])
# 2. what tag do we want to predict?
tag_type = 'upos'
# 3. make the tag dictionary from the corpus
tag_dictionary = corpus.make_tag_dictionary(tag_type=tag_type)
# 4. initialize each embedding we use
embedding_types = [
# contextual string embeddings, forward
FlairEmbeddings('multi-forward'),
# contextual string embeddings, backward
FlairEmbeddings('multi-backward'),
]
# embedding stack consists of Flair and GloVe embeddings
embeddings = StackedEmbeddings(embeddings=embedding_types)
# 5. initialize sequence tagger
from flair.models import SequenceTagger
tagger = SequenceTagger(hidden_size=256,
embeddings=embeddings,
tag_dictionary=tag_dictionary,
tag_type=tag_type,
use_crf=False)
# 6. initialize trainer
from flair.trainers import ModelTrainer
trainer = ModelTrainer(tagger, corpus)
# 7. run training
trainer.train('resources/taggers/upos-multi',
train_with_dev=True,
max_epochs=150)
```
---
### Cite
Please cite the following paper when using this model.
```
@inproceedings{akbik2018coling,
title={Contextual String Embeddings for Sequence Labeling},
author={Akbik, Alan and Blythe, Duncan and Vollgraf, Roland},
booktitle = {{COLING} 2018, 27th International Conference on Computational Linguistics},
pages = {1638--1649},
year = {2018}
}
```
---
### Issues?
The Flair issue tracker is available [here](https://github.com/flairNLP/flair/issues/).
| 4,755 | [
[
-0.0306854248046875,
-0.0384521484375,
0.01015472412109375,
0.01020050048828125,
-0.022674560546875,
0.00044083595275878906,
-0.024017333984375,
-0.0287933349609375,
0.037750244140625,
0.01371002197265625,
-0.03363037109375,
-0.051025390625,
-0.02874755859375,
0.020721435546875,
0.0100555419921875,
0.08538818359375,
0.00485992431640625,
0.032806396484375,
-0.0026340484619140625,
-0.0145111083984375,
-0.0316162109375,
-0.05792236328125,
-0.029144287109375,
-0.01267242431640625,
0.0361328125,
0.022186279296875,
0.04010009765625,
0.048370361328125,
0.01678466796875,
0.02362060546875,
-0.0179595947265625,
0.008331298828125,
-0.005786895751953125,
-0.0011425018310546875,
-0.0110015869140625,
-0.0263824462890625,
-0.046722412109375,
0.005115509033203125,
0.051239013671875,
0.04107666015625,
0.00728607177734375,
0.01000213623046875,
-0.0017290115356445312,
0.00876617431640625,
-0.02630615234375,
0.0286407470703125,
-0.046875,
-0.018768310546875,
-0.02276611328125,
0.000789642333984375,
-0.0310211181640625,
-0.00868988037109375,
-0.003040313720703125,
-0.035003662109375,
0.00893402099609375,
0.00821685791015625,
0.0887451171875,
0.0092010498046875,
-0.031768798828125,
-0.01486968994140625,
-0.037994384765625,
0.062042236328125,
-0.06396484375,
0.027313232421875,
0.0214385986328125,
-0.011749267578125,
-0.0096282958984375,
-0.041107177734375,
-0.04632568359375,
-0.0108184814453125,
-0.0118865966796875,
0.0218963623046875,
-0.009979248046875,
-0.006885528564453125,
0.0133819580078125,
0.00960540771484375,
-0.050048828125,
-0.0000028014183044433594,
-0.01216888427734375,
-0.027435302734375,
0.0518798828125,
0.00347137451171875,
0.01678466796875,
-0.032989501953125,
-0.040374755859375,
-0.013763427734375,
-0.026947021484375,
0.01070404052734375,
0.01122283935546875,
0.04693603515625,
-0.01053619384765625,
0.0380859375,
0.007904052734375,
0.052398681640625,
-0.0011768341064453125,
-0.024932861328125,
0.050323486328125,
-0.03546142578125,
-0.01462554931640625,
-0.0035114288330078125,
0.07708740234375,
0.0181427001953125,
0.01380157470703125,
-0.003429412841796875,
-0.004299163818359375,
0.0155181884765625,
-0.0190887451171875,
-0.03656005859375,
-0.011566162109375,
0.0233306884765625,
-0.01332855224609375,
-0.0105438232421875,
0.00823974609375,
-0.058929443359375,
-0.01287078857421875,
-0.0111541748046875,
0.033935546875,
-0.044647216796875,
-0.014984130859375,
0.01092529296875,
-0.0200347900390625,
0.01517486572265625,
0.0023441314697265625,
-0.055999755859375,
-0.007659912109375,
0.023345947265625,
0.0458984375,
0.01555633544921875,
-0.03314208984375,
-0.0169525146484375,
0.006610870361328125,
-0.0105133056640625,
0.054534912109375,
-0.0352783203125,
-0.0285186767578125,
0.0002639293670654297,
0.016387939453125,
-0.035736083984375,
-0.01186370849609375,
0.05389404296875,
-0.0261993408203125,
0.031341552734375,
-0.0120391845703125,
-0.052886962890625,
-0.0271759033203125,
0.00872802734375,
-0.04486083984375,
0.07403564453125,
-0.0021953582763671875,
-0.080810546875,
0.028045654296875,
-0.035614013671875,
-0.032806396484375,
0.002956390380859375,
-0.00878143310546875,
-0.029815673828125,
-0.01110076904296875,
0.01580810546875,
0.0537109375,
-0.00872802734375,
0.034698486328125,
-0.0262603759765625,
-0.002674102783203125,
0.014373779296875,
-0.0004858970642089844,
0.072265625,
0.01194000244140625,
-0.0169219970703125,
0.014404296875,
-0.05938720703125,
-0.006565093994140625,
0.01534271240234375,
-0.036346435546875,
-0.0229034423828125,
0.0005855560302734375,
0.01021575927734375,
0.0203399658203125,
0.006778717041015625,
-0.045562744140625,
0.037689208984375,
-0.041748046875,
0.033050537109375,
0.041656494140625,
-0.003570556640625,
0.03680419921875,
-0.0294647216796875,
0.03515625,
0.01543426513671875,
-0.011383056640625,
-0.0185394287109375,
-0.0587158203125,
-0.05352783203125,
-0.037933349609375,
0.03948974609375,
0.057464599609375,
-0.0572509765625,
0.059173583984375,
-0.0298919677734375,
-0.046844482421875,
-0.040252685546875,
-0.01209259033203125,
0.026611328125,
0.043182373046875,
0.0355224609375,
-0.01239776611328125,
-0.06158447265625,
-0.057830810546875,
-0.01275634765625,
-0.0097808837890625,
0.0164642333984375,
0.0084228515625,
0.061676025390625,
-0.015289306640625,
0.06634521484375,
-0.0243682861328125,
-0.0303497314453125,
-0.0236968994140625,
0.0145721435546875,
0.03692626953125,
0.04376220703125,
0.04119873046875,
-0.0518798828125,
-0.050384521484375,
-0.0140533447265625,
-0.0278778076171875,
0.01812744140625,
-0.0120849609375,
0.0036640167236328125,
0.03558349609375,
0.0246734619140625,
-0.043853759765625,
0.026611328125,
0.031707763671875,
-0.04083251953125,
0.046722412109375,
-0.00885772705078125,
-0.01141357421875,
-0.1165771484375,
0.0212860107421875,
0.01316070556640625,
-0.01227569580078125,
-0.04876708984375,
-0.01482391357421875,
0.00440216064453125,
0.01450347900390625,
-0.0389404296875,
0.057464599609375,
-0.031402587890625,
0.00894927978515625,
0.0098724365234375,
-0.0008745193481445312,
0.0013589859008789062,
0.05108642578125,
0.0218963623046875,
0.0452880859375,
0.056182861328125,
-0.047454833984375,
0.0172271728515625,
0.028045654296875,
-0.0247039794921875,
0.005615234375,
-0.0276641845703125,
-0.01169586181640625,
-0.016326904296875,
0.0269622802734375,
-0.09674072265625,
-0.0190277099609375,
0.0379638671875,
-0.058929443359375,
0.04119873046875,
-0.0028324127197265625,
-0.039031982421875,
-0.03570556640625,
-0.0257110595703125,
0.005863189697265625,
0.0229339599609375,
-0.0300140380859375,
0.047027587890625,
0.0333251953125,
0.0029201507568359375,
-0.055145263671875,
-0.051177978515625,
-0.01317596435546875,
-0.0261993408203125,
-0.045562744140625,
0.03582763671875,
-0.010284423828125,
0.005764007568359375,
-0.0020046234130859375,
0.012542724609375,
0.0010318756103515625,
0.01276397705078125,
0.01180267333984375,
0.034149169921875,
-0.01293182373046875,
0.0209503173828125,
-0.017181396484375,
0.0009860992431640625,
-0.01580810546875,
-0.01427459716796875,
0.061981201171875,
-0.0207366943359375,
0.007701873779296875,
-0.0374755859375,
0.0187530517578125,
0.0218048095703125,
-0.033233642578125,
0.056915283203125,
0.06982421875,
-0.0284576416015625,
0.003688812255859375,
-0.0232391357421875,
-0.001377105712890625,
-0.027374267578125,
0.036834716796875,
-0.04632568359375,
-0.057281494140625,
0.047821044921875,
0.0025997161865234375,
0.007518768310546875,
0.06158447265625,
0.04443359375,
-0.007427215576171875,
0.0947265625,
0.043609619140625,
-0.0219573974609375,
0.030731201171875,
-0.03875732421875,
0.00421905517578125,
-0.05645751953125,
-0.0126495361328125,
-0.049102783203125,
-0.00577545166015625,
-0.063720703125,
-0.0303802490234375,
0.0146636962890625,
0.03155517578125,
-0.03253173828125,
0.038177490234375,
-0.03900146484375,
0.017669677734375,
0.05242919921875,
-0.01122283935546875,
0.00682830810546875,
0.001903533935546875,
-0.037994384765625,
-0.008697509765625,
-0.06402587890625,
-0.036834716796875,
0.07275390625,
0.034912109375,
0.044189453125,
0.00101470947265625,
0.060546875,
-0.004848480224609375,
0.019561767578125,
-0.066162109375,
0.0295562744140625,
-0.0217437744140625,
-0.062408447265625,
-0.00841522216796875,
-0.0140228271484375,
-0.0809326171875,
0.0247039794921875,
-0.019775390625,
-0.078369140625,
0.020538330078125,
0.01136016845703125,
-0.038726806640625,
0.03033447265625,
-0.041259765625,
0.07257080078125,
-0.00893402099609375,
-0.01371002197265625,
0.016448974609375,
-0.06396484375,
0.0187225341796875,
0.001331329345703125,
0.043975830078125,
-0.0167388916015625,
-0.0013475418090820312,
0.07867431640625,
-0.00853729248046875,
0.0830078125,
0.0006060600280761719,
0.00742340087890625,
0.0224456787109375,
-0.0015087127685546875,
0.0190887451171875,
0.00664520263671875,
0.0008392333984375,
0.0151519775390625,
0.01076507568359375,
-0.01580810546875,
0.0015859603881835938,
0.04119873046875,
-0.0615234375,
-0.0240478515625,
-0.0650634765625,
-0.02252197265625,
-0.01534271240234375,
0.0242767333984375,
0.05010986328125,
0.028900146484375,
-0.01555633544921875,
-0.0105133056640625,
0.03131103515625,
-0.022125244140625,
0.049896240234375,
0.039459228515625,
-0.02777099609375,
-0.049957275390625,
0.0703125,
-0.0031566619873046875,
-0.00907135009765625,
0.03045654296875,
0.023834228515625,
-0.0341796875,
-0.005199432373046875,
-0.0323486328125,
0.04486083984375,
-0.051361083984375,
-0.033599853515625,
-0.043701171875,
-0.00911712646484375,
-0.06842041015625,
0.0059967041015625,
-0.0243377685546875,
-0.037078857421875,
-0.044677734375,
0.00283050537109375,
0.02056884765625,
0.0472412109375,
-0.0235137939453125,
0.025054931640625,
-0.06280517578125,
-0.0015087127685546875,
-0.004749298095703125,
0.0120697021484375,
-0.0170440673828125,
-0.05963134765625,
-0.0204925537109375,
0.01416778564453125,
-0.0257415771484375,
-0.0794677734375,
0.05987548828125,
0.0301666259765625,
0.04437255859375,
0.0295562744140625,
-0.01296234130859375,
0.0380859375,
-0.034454345703125,
0.068603515625,
0.01885986328125,
-0.07208251953125,
0.03656005859375,
-0.0252685546875,
0.025238037109375,
0.0155487060546875,
0.0679931640625,
-0.05096435546875,
-0.0156707763671875,
-0.054840087890625,
-0.0693359375,
0.0648193359375,
-0.0088043212890625,
0.0080413818359375,
-0.0247344970703125,
0.0093841552734375,
-0.0032711029052734375,
0.00811767578125,
-0.07232666015625,
-0.042022705078125,
-0.01477813720703125,
-0.0170440673828125,
-0.0322265625,
-0.0177001953125,
-0.0021190643310546875,
-0.046844482421875,
0.0782470703125,
-0.01195526123046875,
0.037628173828125,
0.0208892822265625,
0.0009589195251464844,
0.0016880035400390625,
0.0189208984375,
0.0438232421875,
0.0262603759765625,
-0.025421142578125,
-0.001499176025390625,
0.007228851318359375,
-0.01708984375,
-0.00731658935546875,
0.01335906982421875,
-0.0016222000122070312,
0.024017333984375,
0.03375244140625,
0.07110595703125,
0.01470947265625,
-0.0301666259765625,
0.0404052734375,
-0.00284576416015625,
-0.016876220703125,
-0.0302276611328125,
-0.0217437744140625,
0.01763916015625,
0.0081634521484375,
0.009613037109375,
0.01140594482421875,
-0.0085296630859375,
-0.042572021484375,
0.01300811767578125,
0.0295562744140625,
-0.0255889892578125,
-0.0384521484375,
0.058258056640625,
0.0047149658203125,
-0.00525665283203125,
0.0266571044921875,
-0.031036376953125,
-0.06842041015625,
0.03948974609375,
0.043548583984375,
0.05389404296875,
-0.033447265625,
0.0186920166015625,
0.059173583984375,
0.00466156005859375,
-0.003261566162109375,
0.048675537109375,
0.0213623046875,
-0.07684326171875,
-0.03271484375,
-0.071044921875,
0.006664276123046875,
0.0141754150390625,
-0.042327880859375,
0.0297698974609375,
-0.0176849365234375,
-0.03228759765625,
0.0288543701171875,
0.004116058349609375,
-0.060302734375,
0.01374053955078125,
0.038726806640625,
0.07135009765625,
-0.075927734375,
0.09027099609375,
0.0797119140625,
-0.059783935546875,
-0.0699462890625,
-0.0202178955078125,
-0.0018606185913085938,
-0.05755615234375,
0.0518798828125,
0.0243377685546875,
0.0184783935546875,
0.0101470947265625,
-0.0457763671875,
-0.0924072265625,
0.06964111328125,
-0.0024662017822265625,
-0.02752685546875,
-0.01000213623046875,
0.0003306865692138672,
0.039398193359375,
-0.034912109375,
0.0243682861328125,
0.0389404296875,
0.034332275390625,
0.00872802734375,
-0.080810546875,
-0.0069427490234375,
-0.0157470703125,
-0.0005326271057128906,
0.00482177734375,
-0.04815673828125,
0.08428955078125,
-0.01540374755859375,
-0.023651123046875,
0.016326904296875,
0.056884765625,
0.004878997802734375,
-0.005954742431640625,
0.02630615234375,
0.052337646484375,
0.05999755859375,
-0.01419830322265625,
0.06854248046875,
-0.03619384765625,
0.0455322265625,
0.0943603515625,
-0.00044035911560058594,
0.0684814453125,
0.0286407470703125,
-0.01047515869140625,
0.033905029296875,
0.06536865234375,
-0.01146697998046875,
0.032684326171875,
0.0069580078125,
-0.0064697265625,
-0.0283966064453125,
-0.02069091796875,
-0.02874755859375,
0.057586669921875,
0.0302734375,
-0.0287933349609375,
-0.003261566162109375,
0.011322021484375,
0.035675048828125,
-0.00998687744140625,
-0.016082763671875,
0.056121826171875,
-0.00875091552734375,
-0.0491943359375,
0.0574951171875,
0.014678955078125,
0.0684814453125,
-0.043060302734375,
0.00907135009765625,
-0.00701141357421875,
0.0200347900390625,
-0.005390167236328125,
-0.055145263671875,
0.006061553955078125,
-0.019256591796875,
-0.0008158683776855469,
-0.009521484375,
0.042388916015625,
-0.052886962890625,
-0.0372314453125,
0.0283355712890625,
0.04302978515625,
0.015838623046875,
0.0021114349365234375,
-0.06842041015625,
-0.00757598876953125,
0.0201568603515625,
-0.0300750732421875,
0.01313018798828125,
0.01264190673828125,
0.0001552104949951172,
0.03533935546875,
0.033416748046875,
0.008270263671875,
0.0089569091796875,
-0.0010633468627929688,
0.06756591796875,
-0.0550537109375,
-0.0310211181640625,
-0.060150146484375,
0.05279541015625,
0.0102691650390625,
-0.0345458984375,
0.05621337890625,
0.0526123046875,
0.0762939453125,
-0.0162811279296875,
0.06298828125,
-0.0245513916015625,
0.058746337890625,
-0.0171966552734375,
0.036712646484375,
-0.053680419921875,
-0.0024547576904296875,
-0.01178741455078125,
-0.05401611328125,
-0.035797119140625,
0.043731689453125,
-0.01593017578125,
-0.00853729248046875,
0.043975830078125,
0.05413818359375,
0.01207733154296875,
-0.010101318359375,
0.0141448974609375,
0.0276641845703125,
0.01172637939453125,
0.037261962890625,
0.044525146484375,
-0.033843994140625,
0.0290069580078125,
-0.0457763671875,
-0.0161590576171875,
-0.00714111328125,
-0.060028076171875,
-0.0660400390625,
-0.0662841796875,
-0.0277252197265625,
-0.04168701171875,
-0.0199737548828125,
0.08184814453125,
0.034088134765625,
-0.07763671875,
-0.0271148681640625,
0.0171051025390625,
-0.004932403564453125,
-0.0027332305908203125,
-0.0178375244140625,
0.036834716796875,
-0.028045654296875,
-0.0595703125,
0.022552490234375,
-0.0157470703125,
0.0182037353515625,
0.0048675537109375,
0.0030460357666015625,
-0.060150146484375,
-0.0035648345947265625,
0.032684326171875,
0.02301025390625,
-0.06207275390625,
-0.016143798828125,
0.0013227462768554688,
-0.017608642578125,
0.0177764892578125,
0.0235748291015625,
-0.0478515625,
0.0302734375,
0.0435791015625,
0.01303863525390625,
0.021728515625,
-0.0010662078857421875,
0.0265350341796875,
-0.06561279296875,
0.01473236083984375,
0.025543212890625,
0.04730224609375,
0.02099609375,
-0.00795745849609375,
0.0297088623046875,
0.03497314453125,
-0.039306640625,
-0.043365478515625,
0.0007891654968261719,
-0.0738525390625,
-0.0394287109375,
0.09613037109375,
-0.01450347900390625,
-0.031097412109375,
0.0017557144165039062,
-0.020904541015625,
0.050384521484375,
-0.033447265625,
0.0215301513671875,
0.04541015625,
-0.01202392578125,
0.01885986328125,
-0.0177154541015625,
0.058197021484375,
0.0275726318359375,
-0.0285491943359375,
-0.00553131103515625,
0.02685546875,
0.031646728515625,
0.035675048828125,
0.041778564453125,
0.00855255126953125,
0.0002397298812866211,
-0.00498199462890625,
0.0390625,
0.007595062255859375,
-0.019317626953125,
-0.041778564453125,
-0.01332855224609375,
-0.00518035888671875,
-0.0272979736328125
]
] |
adept/persimmon-8b-chat | 2023-10-11T15:07:27.000Z | [
"transformers",
"pytorch",
"persimmon",
"text-generation",
"license:apache-2.0",
"endpoints_compatible",
"region:us"
] | text-generation | adept | null | null | adept/persimmon-8b-chat | 37 | 7,214 | transformers | 2023-09-07T19:39:24 | ---
license: apache-2.0
---
At Adept, we’re working towards an AI agent that can help people do anything they need to do on a computer. We’re not in the business of shipping isolated language models (LMs)—this was an early output of the model scaling program that will support our products.
We trained it from scratch using a context size of 16K. Many LM use cases are context-bound; our model has 4 times the context size of LLaMA2 and 8 times that of GPT-3, MPT, etc.
This is a chat finetuned version of the base model. The best prompt to use is:
human: [some query]
adept:
See https://www.adept.ai/blog/persimmon-8b for more info | 641 | [
[
-0.0155487060546875,
-0.045166015625,
0.052734375,
0.01177978515625,
-0.0047149658203125,
-0.0020732879638671875,
-0.017181396484375,
-0.04876708984375,
-0.00008857250213623047,
0.047607421875,
-0.036773681640625,
-0.0164337158203125,
-0.03399658203125,
-0.002231597900390625,
-0.0148468017578125,
0.06756591796875,
0.01325225830078125,
-0.0008940696716308594,
-0.00380706787109375,
-0.0213470458984375,
-0.059814453125,
-0.0288848876953125,
-0.08013916015625,
-0.0406494140625,
0.0379638671875,
0.06610107421875,
0.06976318359375,
0.055389404296875,
0.032806396484375,
0.0263824462890625,
0.006885528564453125,
0.003986358642578125,
-0.04327392578125,
0.0176849365234375,
-0.01192474365234375,
-0.03472900390625,
-0.05926513671875,
0.0027179718017578125,
0.033477783203125,
0.044464111328125,
-0.0125274658203125,
0.0144500732421875,
-0.01403045654296875,
0.037017822265625,
-0.01983642578125,
0.01177215576171875,
-0.0399169921875,
-0.0201568603515625,
-0.0130157470703125,
0.0200653076171875,
-0.031097412109375,
0.0203399658203125,
-0.0004954338073730469,
-0.0633544921875,
-0.01369476318359375,
0.00510406494140625,
0.054412841796875,
0.05389404296875,
-0.0406494140625,
0.000988006591796875,
-0.05767822265625,
0.0604248046875,
-0.04901123046875,
0.0269012451171875,
0.04058837890625,
0.047515869140625,
-0.01342010498046875,
-0.072509765625,
-0.04278564453125,
-0.026947021484375,
-0.0020961761474609375,
-0.025482177734375,
-0.02374267578125,
0.0138702392578125,
0.0216064453125,
0.0108642578125,
-0.0226898193359375,
0.016143798828125,
-0.023468017578125,
-0.0209197998046875,
0.056793212890625,
0.0265045166015625,
0.0380859375,
0.020751953125,
-0.01399993896484375,
-0.0024394989013671875,
-0.046112060546875,
-0.0010089874267578125,
-0.00345611572265625,
0.03424072265625,
-0.00962066650390625,
0.043609619140625,
-0.00946044921875,
0.0682373046875,
0.0166168212890625,
-0.01200103759765625,
-0.00737762451171875,
-0.0025577545166015625,
-0.013275146484375,
0.009033203125,
0.04791259765625,
0.01468658447265625,
0.0032329559326171875,
-0.01038360595703125,
-0.034271240234375,
-0.005985260009765625,
0.05645751953125,
-0.040924072265625,
-0.0018930435180664062,
0.01053619384765625,
-0.01013946533203125,
0.00811004638671875,
-0.01474761962890625,
-0.01207733154296875,
-0.043792724609375,
-0.035369873046875,
0.039703369140625,
-0.0228424072265625,
-0.016082763671875,
-0.007701873779296875,
0.01190948486328125,
0.027801513671875,
0.037841796875,
-0.056854248046875,
0.03009033203125,
0.05255126953125,
0.046142578125,
-0.007598876953125,
-0.0207672119140625,
-0.043914794921875,
-0.0023403167724609375,
-0.0176849365234375,
0.062042236328125,
-0.032562255859375,
-0.030303955078125,
-0.0277252197265625,
-0.0005755424499511719,
0.00830841064453125,
-0.0307464599609375,
0.03973388671875,
-0.05352783203125,
-0.016448974609375,
-0.01132965087890625,
-0.036346435546875,
-0.00865936279296875,
0.036529541015625,
-0.049652099609375,
0.06634521484375,
0.032073974609375,
-0.0201873779296875,
0.011016845703125,
-0.06768798828125,
-0.0266571044921875,
0.04107666015625,
-0.0111846923828125,
-0.0153045654296875,
0.0025310516357421875,
0.0099639892578125,
0.03228759765625,
-0.0233612060546875,
0.0011844635009765625,
-0.00878143310546875,
-0.0173187255859375,
0.004230499267578125,
-0.01207733154296875,
0.04180908203125,
0.0355224609375,
-0.017974853515625,
0.0156402587890625,
-0.039031982421875,
0.00412750244140625,
0.0095672607421875,
-0.042877197265625,
-0.009368896484375,
-0.00792694091796875,
-0.0010852813720703125,
-0.019805908203125,
0.05084228515625,
-0.03497314453125,
0.047698974609375,
-0.0167388916015625,
0.0238037109375,
0.042388916015625,
-0.0157928466796875,
0.05010986328125,
-0.0293731689453125,
0.058502197265625,
-0.0166168212890625,
-0.0079803466796875,
-0.004058837890625,
-0.032867431640625,
-0.0511474609375,
-0.0248870849609375,
0.005115509033203125,
0.03765869140625,
-0.050262451171875,
0.01151275634765625,
-0.0296783447265625,
-0.052581787109375,
-0.032806396484375,
0.02239990234375,
0.00732421875,
0.0217437744140625,
0.01474761962890625,
-0.0157318115234375,
-0.039794921875,
-0.0689697265625,
-0.0323486328125,
-0.0248870849609375,
-0.0244598388671875,
0.02972412109375,
0.01171112060546875,
-0.03533935546875,
0.096435546875,
-0.0286712646484375,
-0.036346435546875,
-0.0428466796875,
-0.00963592529296875,
0.0157318115234375,
0.029754638671875,
0.0148468017578125,
-0.0653076171875,
-0.051727294921875,
0.0227813720703125,
-0.056427001953125,
-0.0034313201904296875,
-0.01922607421875,
-0.00806427001953125,
-0.0028400421142578125,
0.04571533203125,
-0.055999755859375,
0.022186279296875,
0.033660888671875,
-0.03570556640625,
0.0210723876953125,
-0.0216522216796875,
-0.0118255615234375,
-0.0858154296875,
-0.00539398193359375,
-0.006439208984375,
-0.04736328125,
-0.055084228515625,
0.01447296142578125,
-0.0007500648498535156,
-0.01788330078125,
-0.04364013671875,
0.050079345703125,
-0.05804443359375,
-0.00963592529296875,
-0.0199432373046875,
0.0037384033203125,
-0.0204010009765625,
0.0235595703125,
-0.00382232666015625,
0.08514404296875,
0.0312347412109375,
-0.0555419921875,
0.0321044921875,
0.0230560302734375,
-0.01557159423828125,
0.059906005859375,
-0.0684814453125,
0.036224365234375,
0.0136260986328125,
0.004299163818359375,
-0.052734375,
-0.045166015625,
-0.0010852813720703125,
-0.038665771484375,
0.038360595703125,
-0.0007467269897460938,
-0.0206451416015625,
-0.0104217529296875,
-0.002803802490234375,
0.0439453125,
0.036712646484375,
-0.044647216796875,
0.06256103515625,
0.058929443359375,
0.0091705322265625,
-0.03192138671875,
-0.036773681640625,
0.047271728515625,
-0.02734375,
-0.059112548828125,
0.01291656494140625,
-0.00936126708984375,
-0.05035400390625,
-0.01544189453125,
0.015960693359375,
-0.01161956787109375,
0.0274505615234375,
0.0233306884765625,
-0.006076812744140625,
-0.0310211181640625,
0.01517486572265625,
-0.0081787109375,
-0.0022144317626953125,
-0.01425933837890625,
-0.01800537109375,
0.02728271484375,
-0.045501708984375,
-0.021575927734375,
-0.046539306640625,
0.0189971923828125,
0.0628662109375,
0.005237579345703125,
0.058837890625,
0.018890380859375,
-0.0236358642578125,
-0.006580352783203125,
-0.025604248046875,
-0.0093841552734375,
-0.039764404296875,
0.0240936279296875,
-0.0239410400390625,
-0.06365966796875,
0.036346435546875,
0.00618743896484375,
0.01311492919921875,
0.029022216796875,
0.044952392578125,
0.013916015625,
0.08026123046875,
0.0770263671875,
-0.03057861328125,
0.038787841796875,
-0.012664794921875,
0.01050567626953125,
-0.06671142578125,
0.013885498046875,
-0.0260162353515625,
-0.039154052734375,
-0.00914764404296875,
-0.0299224853515625,
0.056365966796875,
0.0033168792724609375,
-0.034820556640625,
0.0518798828125,
-0.046417236328125,
0.038543701171875,
0.04595947265625,
0.00791168212890625,
0.021392822265625,
-0.0297698974609375,
0.024169921875,
0.0159912109375,
-0.059722900390625,
-0.06927490234375,
0.08050537109375,
0.039520263671875,
0.058990478515625,
0.03057861328125,
0.030517578125,
0.044830322265625,
0.045166015625,
-0.083984375,
0.0518798828125,
0.023406982421875,
-0.06292724609375,
-0.04583740234375,
-0.009185791015625,
-0.088134765625,
-0.010467529296875,
0.00830078125,
-0.06298828125,
-0.014923095703125,
0.01422882080078125,
-0.030517578125,
0.02557373046875,
-0.0406494140625,
0.06890869140625,
-0.0205535888671875,
-0.01861572265625,
-0.014373779296875,
-0.0650634765625,
0.01096343994140625,
0.0095672607421875,
0.004787445068359375,
-0.0134735107421875,
-0.03155517578125,
0.047454833984375,
-0.055023193359375,
0.08123779296875,
-0.0011205673217773438,
-0.015350341796875,
0.02630615234375,
0.02154541015625,
0.03057861328125,
-0.0080718994140625,
-0.00936126708984375,
0.006725311279296875,
-0.004230499267578125,
-0.036407470703125,
-0.0252685546875,
0.0460205078125,
-0.0833740234375,
-0.056884765625,
-0.015869140625,
-0.05419921875,
-0.0097198486328125,
0.0242919921875,
0.0156402587890625,
0.023468017578125,
-0.033294677734375,
0.01087188720703125,
0.032257080078125,
0.01360321044921875,
0.033050537109375,
0.04779052734375,
0.00797271728515625,
-0.0028171539306640625,
0.049468994140625,
-0.0013723373413085938,
0.0030345916748046875,
0.033477783203125,
0.0113372802734375,
-0.0081634521484375,
-0.0240936279296875,
-0.05718994140625,
0.01444244384765625,
-0.052032470703125,
-0.0267181396484375,
-0.037841796875,
0.00118255615234375,
-0.037567138671875,
-0.0213775634765625,
-0.04107666015625,
-0.035064697265625,
-0.048797607421875,
-0.00234222412109375,
0.0321044921875,
0.080078125,
0.0012359619140625,
0.060394287109375,
-0.06640625,
-0.0048980712890625,
0.03216552734375,
0.003253936767578125,
-0.0073699951171875,
-0.042724609375,
-0.030303955078125,
0.006229400634765625,
-0.043914794921875,
-0.04937744140625,
0.0267181396484375,
0.037994384765625,
0.0523681640625,
0.042694091796875,
0.024383544921875,
0.0225982666015625,
-0.031402587890625,
0.088134765625,
-0.01413726806640625,
-0.06365966796875,
0.0222320556640625,
-0.0467529296875,
0.04620361328125,
0.03961181640625,
0.031402587890625,
-0.057220458984375,
-0.032958984375,
-0.042083740234375,
-0.049774169921875,
0.052764892578125,
-0.008270263671875,
0.0341796875,
-0.0004968643188476562,
0.047454833984375,
-0.0023021697998046875,
0.0018415451049804688,
-0.04345703125,
-0.006755828857421875,
0.00502777099609375,
-0.02349853515625,
-0.0201873779296875,
-0.0214385986328125,
-0.007663726806640625,
-0.0108489990234375,
0.050537109375,
-0.02508544921875,
0.020660400390625,
0.0120086669921875,
-0.0284423828125,
-0.01003265380859375,
0.0107269287109375,
0.045013427734375,
0.05224609375,
0.0185699462890625,
0.0172119140625,
0.02227783203125,
-0.044097900390625,
0.0120849609375,
-0.020904541015625,
0.00579833984375,
-0.037017822265625,
0.042724609375,
0.056854248046875,
-0.002811431884765625,
-0.06610107421875,
0.021453857421875,
-0.015655517578125,
-0.0011968612670898438,
-0.046844482421875,
0.0250091552734375,
0.01016998291015625,
0.0298919677734375,
0.026580810546875,
-0.018951416015625,
0.0018329620361328125,
-0.0321044921875,
0.007701873779296875,
0.0256195068359375,
-0.050872802734375,
-0.038848876953125,
0.06707763671875,
0.02691650390625,
-0.068603515625,
0.06317138671875,
-0.005199432373046875,
-0.05419921875,
0.041839599609375,
0.03173828125,
0.050537109375,
-0.006877899169921875,
0.0020999908447265625,
0.007389068603515625,
0.02569580078125,
-0.0166168212890625,
-0.00539398193359375,
-0.0240936279296875,
-0.06915283203125,
-0.0239410400390625,
-0.025054931640625,
-0.0419921875,
-0.01062774658203125,
-0.036529541015625,
0.0264739990234375,
-0.037139892578125,
-0.025177001953125,
-0.0130767822265625,
0.01125335693359375,
-0.03466796875,
-0.0000368952751159668,
0.0017499923706054688,
0.08203125,
-0.03887939453125,
0.0634765625,
0.0906982421875,
-0.0276031494140625,
-0.0855712890625,
-0.0192718505859375,
0.0152587890625,
-0.06768798828125,
0.035308837890625,
0.01275634765625,
-0.0040740966796875,
0.01045989990234375,
-0.0689697265625,
-0.05096435546875,
0.08795166015625,
0.0202484130859375,
-0.0193634033203125,
0.0119781494140625,
0.004924774169921875,
0.053314208984375,
-0.040283203125,
0.00716400146484375,
0.06256103515625,
0.043487548828125,
-0.0011224746704101562,
-0.0860595703125,
-0.0028171539306640625,
-0.025238037109375,
-0.0083160400390625,
0.00799560546875,
-0.03814697265625,
0.069091796875,
-0.0193328857421875,
-0.005588531494140625,
0.025421142578125,
0.05999755859375,
-0.01314544677734375,
0.0167388916015625,
0.050262451171875,
0.042083740234375,
0.060211181640625,
0.016326904296875,
0.0987548828125,
-0.022216796875,
0.007755279541015625,
0.08929443359375,
-0.021514892578125,
0.05029296875,
0.032073974609375,
-0.00574493408203125,
0.037384033203125,
0.05157470703125,
0.0309295654296875,
0.008636474609375,
-0.00942230224609375,
-0.037628173828125,
-0.0294647216796875,
-0.014373779296875,
-0.00351715087890625,
0.0325927734375,
0.0267486572265625,
0.0048980712890625,
-0.004673004150390625,
0.00475311279296875,
0.00914764404296875,
0.021575927734375,
-0.0283050537109375,
0.08026123046875,
0.00884246826171875,
-0.05572509765625,
0.0455322265625,
0.00611114501953125,
0.0238189697265625,
-0.047332763671875,
0.00884246826171875,
-0.023712158203125,
0.041595458984375,
0.000011086463928222656,
-0.04852294921875,
0.0266265869140625,
-0.01451873779296875,
-0.000011980533599853516,
-0.015838623046875,
0.046905517578125,
-0.0224151611328125,
-0.0198822021484375,
-0.0013027191162109375,
0.021575927734375,
0.042510986328125,
-0.0251312255859375,
-0.040435791015625,
0.0060272216796875,
-0.0121002197265625,
-0.023956298828125,
0.0187835693359375,
0.039520263671875,
-0.0297698974609375,
0.0797119140625,
0.0157928466796875,
-0.004360198974609375,
-0.01531982421875,
-0.0268096923828125,
0.0679931640625,
-0.02825927734375,
-0.031280517578125,
-0.04290771484375,
0.036956787109375,
-0.0011577606201171875,
-0.026123046875,
0.072021484375,
0.029327392578125,
0.03680419921875,
0.01087188720703125,
0.034423828125,
-0.01192474365234375,
0.033599853515625,
0.0020771026611328125,
0.05401611328125,
-0.06317138671875,
0.035675048828125,
0.01042938232421875,
-0.05401611328125,
-0.0253448486328125,
0.040069580078125,
-0.0213623046875,
-0.00762939453125,
0.022705078125,
0.05615234375,
-0.00873565673828125,
-0.0097198486328125,
0.050262451171875,
0.024169921875,
0.03790283203125,
0.0261993408203125,
0.0626220703125,
-0.00151824951171875,
0.0257110595703125,
0.01320648193359375,
-0.02911376953125,
-0.0289459228515625,
-0.06634521484375,
-0.0594482421875,
-0.031341552734375,
-0.004833221435546875,
-0.019805908203125,
-0.017974853515625,
0.082763671875,
0.06854248046875,
-0.035186767578125,
-0.037994384765625,
0.0038318634033203125,
0.0167388916015625,
-0.0160369873046875,
-0.007686614990234375,
-0.01232147216796875,
-0.02313232421875,
-0.0498046875,
0.054443359375,
-0.00264739990234375,
0.00405120849609375,
-0.037445068359375,
-0.0236053466796875,
-0.0100555419921875,
0.010955810546875,
0.033355712890625,
0.0440673828125,
-0.02044677734375,
-0.043609619140625,
-0.00042247772216796875,
-0.00732421875,
0.00246429443359375,
0.06475830078125,
-0.0450439453125,
0.018829345703125,
0.0113372802734375,
0.056304931640625,
0.0285186767578125,
-0.0009794235229492188,
0.0714111328125,
-0.052825927734375,
0.0284881591796875,
0.031097412109375,
0.0122222900390625,
0.033294677734375,
-0.058013916015625,
0.04327392578125,
-0.017852783203125,
-0.06378173828125,
-0.045623779296875,
0.0184783935546875,
-0.07745361328125,
-0.0247039794921875,
0.088134765625,
-0.0012569427490234375,
-0.031585693359375,
-0.01384735107421875,
-0.043548583984375,
-0.0003764629364013672,
-0.0390625,
0.033355712890625,
0.03631591796875,
-0.01094818115234375,
-0.00010836124420166016,
-0.0238494873046875,
0.022613525390625,
0.01629638671875,
-0.043304443359375,
0.0019254684448242188,
0.037841796875,
0.019866943359375,
0.00879669189453125,
0.06512451171875,
0.023651123046875,
0.0180206298828125,
0.00917816162109375,
0.01287078857421875,
-0.01502227783203125,
-0.0247039794921875,
-0.0285186767578125,
-0.0079193115234375,
0.039031982421875,
-0.0009870529174804688
]
] |
AtAndDev/ShortKing-1.4b-v0.1 | 2023-09-29T20:30:08.000Z | [
"transformers",
"pytorch",
"safetensors",
"gpt_neox",
"text-generation",
"en",
"dataset:vicgalle/alpaca-gpt4",
"license:cc-by-nc-4.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | AtAndDev | null | null | AtAndDev/ShortKing-1.4b-v0.1 | 2 | 7,209 | transformers | 2023-09-25T20:26:25 | ---
license: cc-by-nc-4.0
datasets:
- vicgalle/alpaca-gpt4
language:
- en
---
## Model Overview
Model license: cc-by-nc-4.0<br>
This model is trained based on [EleutherAI/pythia-1.4b-deduped](https://huggingface.co/EleutherAI/pythia-1.4b-deduped) model that is LoRA finetuned on [vicgalle/alpaca-gpt4](https://huggingface.co/datasets/vicgalle/alpaca-gpt4) dataset.<br>
## Prompt Template: `Alpaca`
```
<system_prompt>
### Instruction:
<user_message>
### Response:
<assistant_response>
```
## Intended Use
THIS IS A TEST MODEL, IT IS NOT INTENDED FOR REAL APPLICATIONS BY ANY MEANS. HOWEVER, A NEW MODEL IS COMING IN THE SAME TOPIC.<br>
This model series will be used for small but intense applications.
## Training Details
This model took `2:31:23` to train in QLoRA on a single `T4` GPU.<br>
- *epochs*: `1`
- *train batch size*: `12`
- *eval batch size*: `12`
- *gradient accumulation steps*: `1`
- *maximum gradient normal*: `0.3`
- *learning rate*: `2e-4`
- *weight decay*: `0.001`
- *optimizer*: `paged_adamw_32bit`
- *learning rate schedule*: `cosine`
- *warmup ratio (linear)*: `0.03` | 1,119 | [
[
-0.03704833984375,
-0.053466796875,
0.0200653076171875,
0.0018157958984375,
-0.051025390625,
-0.032379150390625,
0.007015228271484375,
-0.043243408203125,
0.0237579345703125,
0.02813720703125,
-0.054779052734375,
-0.033294677734375,
-0.034149169921875,
-0.013916015625,
-0.0195465087890625,
0.08673095703125,
-0.011688232421875,
0.003917694091796875,
0.0291900634765625,
-0.032440185546875,
-0.0257415771484375,
-0.03253173828125,
-0.07183837890625,
-0.035736083984375,
0.055511474609375,
0.01593017578125,
0.04876708984375,
0.03668212890625,
0.0262298583984375,
0.016265869140625,
-0.00920867919921875,
0.0048675537109375,
-0.042022705078125,
-0.0318603515625,
0.019775390625,
-0.0239410400390625,
-0.054290771484375,
0.01084136962890625,
0.04852294921875,
0.0160369873046875,
-0.0250396728515625,
0.0310821533203125,
0.001628875732421875,
0.034271240234375,
-0.046630859375,
0.01476287841796875,
-0.0308685302734375,
0.022247314453125,
-0.01197052001953125,
-0.0014257431030273438,
-0.0023212432861328125,
-0.0367431640625,
0.0187225341796875,
-0.06976318359375,
0.0321044921875,
-0.013519287109375,
0.0826416015625,
0.0179290771484375,
-0.03375244140625,
-0.00457000732421875,
-0.043670654296875,
0.0400390625,
-0.06158447265625,
0.01497650146484375,
0.02752685546875,
0.037994384765625,
0.0189208984375,
-0.056671142578125,
-0.0477294921875,
0.0049285888671875,
0.01537322998046875,
-0.006389617919921875,
-0.0188751220703125,
-0.00972747802734375,
0.03289794921875,
0.033660888671875,
-0.0343017578125,
0.0162506103515625,
-0.0390625,
-0.005146026611328125,
0.029022216796875,
0.0174407958984375,
0.0018758773803710938,
0.0019063949584960938,
-0.042236328125,
-0.0191497802734375,
-0.05084228515625,
0.01299285888671875,
0.032989501953125,
0.021514892578125,
-0.044769287109375,
0.044677734375,
-0.0171051025390625,
0.04193115234375,
-0.004444122314453125,
-0.0123291015625,
0.050201416015625,
-0.003284454345703125,
-0.0546875,
-0.004329681396484375,
0.061248779296875,
0.01629638671875,
-0.01194000244140625,
0.018157958984375,
-0.0093231201171875,
0.0004608631134033203,
0.0035381317138671875,
-0.07513427734375,
-0.0312347412109375,
0.0104217529296875,
-0.0245361328125,
-0.038818359375,
0.0079345703125,
-0.04534912109375,
-0.0173492431640625,
-0.0167388916015625,
0.039642333984375,
-0.0300140380859375,
-0.0184173583984375,
0.01593017578125,
0.0231475830078125,
0.04754638671875,
0.03759765625,
-0.05926513671875,
0.03436279296875,
0.02752685546875,
0.06732177734375,
0.0008077621459960938,
-0.026031494140625,
-0.03192138671875,
-0.007686614990234375,
0.0024967193603515625,
0.041748046875,
0.0220947265625,
-0.019775390625,
-0.0133056640625,
0.0239410400390625,
-0.01031494140625,
-0.0200958251953125,
0.059906005859375,
-0.047607421875,
0.017578125,
-0.036590576171875,
-0.0258026123046875,
-0.04248046875,
0.0183563232421875,
-0.0704345703125,
0.059417724609375,
0.032501220703125,
-0.06884765625,
0.0217742919921875,
-0.043182373046875,
0.013092041015625,
0.0053253173828125,
0.00736236572265625,
-0.07061767578125,
-0.017242431640625,
0.004039764404296875,
0.0106353759765625,
-0.02801513671875,
-0.0084075927734375,
-0.0162353515625,
-0.041839599609375,
-0.00351715087890625,
-0.03814697265625,
0.059967041015625,
0.01403045654296875,
-0.03448486328125,
0.0081939697265625,
-0.0654296875,
0.006900787353515625,
0.03045654296875,
-0.044769287109375,
0.00827789306640625,
-0.035247802734375,
0.0234832763671875,
0.00983428955078125,
0.024383544921875,
-0.037841796875,
0.032562255859375,
-0.018951416015625,
0.029632568359375,
0.05010986328125,
0.00821685791015625,
0.0173492431640625,
-0.0283355712890625,
0.0474853515625,
0.0021820068359375,
0.035797119140625,
0.0279083251953125,
-0.05072021484375,
-0.0714111328125,
-0.02471923828125,
0.018798828125,
0.034637451171875,
-0.044921875,
0.026580810546875,
0.001728057861328125,
-0.049591064453125,
-0.0170135498046875,
0.000957489013671875,
0.021881103515625,
0.045440673828125,
0.033843994140625,
-0.0245513916015625,
-0.033416748046875,
-0.059600830078125,
0.0242156982421875,
0.0004165172576904297,
0.018218994140625,
0.0009059906005859375,
0.057281494140625,
-0.0164031982421875,
0.04534912109375,
-0.042327880859375,
-0.01036834716796875,
-0.00876617431640625,
0.016632080078125,
0.030364990234375,
0.034637451171875,
0.05755615234375,
-0.03515625,
-0.02081298828125,
-0.0177459716796875,
-0.056182861328125,
0.00634002685546875,
0.01404571533203125,
-0.03277587890625,
0.0148773193359375,
-0.003658294677734375,
-0.06268310546875,
0.056396484375,
0.05206298828125,
-0.03814697265625,
0.048431396484375,
-0.0196380615234375,
0.0011739730834960938,
-0.06927490234375,
0.01181793212890625,
0.007404327392578125,
-0.01251983642578125,
-0.0281982421875,
0.017578125,
0.0022640228271484375,
-0.002292633056640625,
-0.05706787109375,
0.042999267578125,
-0.0273590087890625,
-0.0132904052734375,
-0.03363037109375,
-0.01386260986328125,
0.00826263427734375,
0.06024169921875,
-0.007205963134765625,
0.043365478515625,
0.03094482421875,
-0.049591064453125,
0.036529541015625,
0.032257080078125,
-0.0296173095703125,
0.0089263916015625,
-0.0758056640625,
0.014556884765625,
0.0216522216796875,
0.023040771484375,
-0.0276336669921875,
-0.007328033447265625,
0.045928955078125,
-0.0104217529296875,
0.0081329345703125,
-0.0113677978515625,
-0.0279083251953125,
-0.038482666015625,
-0.0295562744140625,
0.05322265625,
0.039520263671875,
-0.0570068359375,
0.039520263671875,
-0.0020732879638671875,
0.0239410400390625,
-0.035491943359375,
-0.037200927734375,
-0.04425048828125,
-0.036468505859375,
-0.03173828125,
0.01959228515625,
-0.00724029541015625,
-0.010955810546875,
-0.016082763671875,
0.015289306640625,
-0.00510406494140625,
0.00347137451171875,
0.0262298583984375,
0.033599853515625,
-0.006378173828125,
-0.005207061767578125,
-0.004528045654296875,
-0.00125885009765625,
-0.004459381103515625,
-0.005527496337890625,
0.06494140625,
-0.01181793212890625,
-0.0283203125,
-0.0675048828125,
-0.0036029815673828125,
0.032470703125,
-0.022491455078125,
0.07672119140625,
0.06158447265625,
-0.042144775390625,
0.013427734375,
-0.02484130859375,
-0.00954437255859375,
-0.03131103515625,
0.032867431640625,
-0.0287628173828125,
-0.02093505859375,
0.058135986328125,
0.03546142578125,
0.0032405853271484375,
0.0628662109375,
0.033782958984375,
0.0237579345703125,
0.061981201171875,
0.04132080078125,
-0.0130157470703125,
0.043365478515625,
-0.06414794921875,
-0.00955963134765625,
-0.0755615234375,
-0.020599365234375,
-0.025421142578125,
-0.00994873046875,
-0.04986572265625,
-0.028076171875,
0.030914306640625,
0.0213165283203125,
-0.06671142578125,
0.0220184326171875,
-0.0298919677734375,
0.00986480712890625,
0.043426513671875,
0.04345703125,
0.01763916015625,
0.01161956787109375,
0.020660400390625,
0.0162200927734375,
-0.055450439453125,
-0.0280303955078125,
0.0838623046875,
0.049041748046875,
0.04498291015625,
0.01525115966796875,
0.048065185546875,
0.00555419921875,
0.0215606689453125,
-0.053192138671875,
0.02056884765625,
0.0166168212890625,
-0.02618408203125,
-0.0232391357421875,
-0.03375244140625,
-0.0682373046875,
0.0235443115234375,
-0.0279541015625,
-0.033538818359375,
0.0217742919921875,
0.01499176025390625,
-0.05291748046875,
0.043975830078125,
-0.0288238525390625,
0.06390380859375,
-0.0131378173828125,
-0.037841796875,
0.0037021636962890625,
-0.0180206298828125,
0.032379150390625,
-0.000659942626953125,
-0.0008463859558105469,
-0.0030536651611328125,
0.01605224609375,
0.0777587890625,
-0.0667724609375,
0.0535888671875,
-0.034576416015625,
-0.034393310546875,
0.04736328125,
-0.01319122314453125,
0.0576171875,
0.008087158203125,
-0.0052947998046875,
0.01415252685546875,
-0.006404876708984375,
-0.0521240234375,
-0.01432037353515625,
0.057861328125,
-0.095703125,
-0.0103912353515625,
-0.058319091796875,
-0.03521728515625,
0.0013599395751953125,
0.0028362274169921875,
0.05023193359375,
0.042877197265625,
-0.0151519775390625,
-0.02069091796875,
0.043792724609375,
-0.01216888427734375,
0.0281982421875,
0.02850341796875,
-0.015350341796875,
-0.036224365234375,
0.05206298828125,
0.00013124942779541016,
0.01485443115234375,
-0.01114654541015625,
0.0009317398071289062,
-0.01446533203125,
-0.0419921875,
-0.038177490234375,
0.046722412109375,
-0.050628662109375,
-0.0170745849609375,
-0.033477783203125,
-0.0168304443359375,
-0.01898193359375,
0.01824951171875,
-0.0340576171875,
-0.03289794921875,
-0.047576904296875,
-0.01995849609375,
0.046356201171875,
0.053314208984375,
0.01189422607421875,
0.056884765625,
-0.05035400390625,
0.00020301342010498047,
0.033355712890625,
0.00257110595703125,
0.0022563934326171875,
-0.07269287109375,
-0.035400390625,
0.0162353515625,
-0.045196533203125,
-0.0654296875,
0.052032470703125,
0.00641632080078125,
0.0261077880859375,
0.0295257568359375,
-0.024169921875,
0.0372314453125,
-0.0148468017578125,
0.053955078125,
0.028076171875,
-0.042755126953125,
0.032440185546875,
-0.046875,
0.0177001953125,
0.0295562744140625,
0.0289306640625,
0.00836944580078125,
0.016265869140625,
-0.06646728515625,
-0.063232421875,
0.042449951171875,
0.0124969482421875,
-0.01125335693359375,
0.0223388671875,
0.04632568359375,
0.0102691650390625,
0.01284027099609375,
-0.07745361328125,
-0.0187835693359375,
-0.026458740234375,
0.0016336441040039062,
-0.013946533203125,
-0.0121917724609375,
-0.01097869873046875,
-0.033660888671875,
0.078857421875,
-0.0133056640625,
0.02923583984375,
0.0018100738525390625,
0.006366729736328125,
-0.0152130126953125,
-0.011505126953125,
0.036834716796875,
0.04083251953125,
-0.04241943359375,
-0.025848388671875,
0.01396942138671875,
-0.044342041015625,
0.002574920654296875,
0.0204620361328125,
-0.021270751953125,
-0.0164794921875,
0.018890380859375,
0.0858154296875,
-0.00045299530029296875,
-0.00820159912109375,
0.0171661376953125,
-0.01393890380859375,
-0.0239715576171875,
-0.0391845703125,
0.01016998291015625,
-0.00475311279296875,
0.0250396728515625,
0.01091766357421875,
-0.002452850341796875,
0.0044403076171875,
-0.043670654296875,
-0.019287109375,
0.021209716796875,
-0.0015964508056640625,
-0.0267791748046875,
0.047515869140625,
0.002040863037109375,
-0.0015411376953125,
0.05804443359375,
-0.0297393798828125,
-0.032318115234375,
0.06622314453125,
0.03411865234375,
0.039886474609375,
-0.021270751953125,
0.00731658935546875,
0.057647705078125,
-0.004123687744140625,
-0.02911376953125,
0.039031982421875,
0.0211334228515625,
-0.053314208984375,
0.002193450927734375,
-0.0467529296875,
-0.02423095703125,
0.036529541015625,
-0.0740966796875,
0.0312347412109375,
-0.0394287109375,
-0.0286407470703125,
-0.008880615234375,
0.030731201171875,
-0.064697265625,
0.03778076171875,
0.0143890380859375,
0.06591796875,
-0.07647705078125,
0.08245849609375,
0.0239410400390625,
-0.05084228515625,
-0.079833984375,
-0.02374267578125,
-0.0177001953125,
-0.0570068359375,
0.0293121337890625,
0.005435943603515625,
-0.006931304931640625,
0.0023899078369140625,
-0.05023193359375,
-0.06475830078125,
0.126953125,
0.02752685546875,
-0.046722412109375,
-0.0106964111328125,
-0.0162353515625,
0.030914306640625,
-0.0222015380859375,
0.025115966796875,
0.039154052734375,
0.012420654296875,
0.00771331787109375,
-0.061004638671875,
-0.003582000732421875,
-0.02679443359375,
-0.0021800994873046875,
0.0020236968994140625,
-0.07635498046875,
0.11334228515625,
-0.01409149169921875,
0.0108184814453125,
0.03753662109375,
0.05938720703125,
0.044525146484375,
0.0168914794921875,
0.0306243896484375,
0.07196044921875,
0.064453125,
-0.00616455078125,
0.0860595703125,
-0.0151519775390625,
0.054534912109375,
0.089111328125,
-0.0109710693359375,
0.05419921875,
0.0240478515625,
-0.0218963623046875,
0.048980712890625,
0.068603515625,
-0.00958251953125,
0.04241943359375,
0.005344390869140625,
-0.0286865234375,
0.00440216064453125,
0.00624847412109375,
-0.05755615234375,
0.02484130859375,
0.0106964111328125,
-0.03289794921875,
0.00011408329010009766,
0.00453948974609375,
0.0012674331665039062,
-0.033477783203125,
-0.038238525390625,
0.01397705078125,
-0.004749298095703125,
-0.03570556640625,
0.07098388671875,
0.013458251953125,
0.056793212890625,
-0.051239013671875,
0.0005464553833007812,
-0.0174713134765625,
0.0146331787109375,
-0.0179901123046875,
-0.00799560546875,
-0.01381683349609375,
-0.00787353515625,
-0.006633758544921875,
0.033599853515625,
0.0298004150390625,
-0.010101318359375,
-0.048004150390625,
0.0157318115234375,
0.036163330078125,
0.0114593505859375,
-0.0028476715087890625,
-0.05816650390625,
0.01215362548828125,
-0.001163482666015625,
-0.0394287109375,
0.025543212890625,
0.01145172119140625,
0.0002008676528930664,
0.051025390625,
0.0372314453125,
0.00133514404296875,
0.0157623291015625,
0.0240020751953125,
0.08135986328125,
-0.03277587890625,
-0.038482666015625,
-0.04266357421875,
0.0062103271484375,
0.01140594482421875,
-0.042236328125,
0.05804443359375,
0.046875,
0.04351806640625,
0.0016384124755859375,
0.05059814453125,
0.01168060302734375,
0.01947021484375,
-0.0611572265625,
0.029693603515625,
-0.0274658203125,
0.00969696044921875,
-0.0131683349609375,
-0.0706787109375,
0.020660400390625,
0.0767822265625,
-0.017486572265625,
0.027557373046875,
0.05352783203125,
0.0413818359375,
-0.016693115234375,
0.0228424072265625,
-0.004848480224609375,
0.0127410888671875,
0.0195159912109375,
0.06494140625,
0.037078857421875,
-0.0745849609375,
0.01058197021484375,
-0.03997802734375,
-0.0155029296875,
-0.00818634033203125,
-0.046783447265625,
-0.04962158203125,
-0.0114898681640625,
-0.034576416015625,
-0.0369873046875,
-0.0084381103515625,
0.07147216796875,
0.0460205078125,
-0.04962158203125,
-0.0181427001953125,
-0.015869140625,
-0.02825927734375,
-0.0154266357421875,
-0.01291656494140625,
0.049560546875,
0.0292816162109375,
-0.059722900390625,
0.00577545166015625,
-0.03778076171875,
0.0452880859375,
-0.0292816162109375,
-0.02459716796875,
-0.0289306640625,
-0.01175689697265625,
0.017822265625,
0.039794921875,
-0.0231170654296875,
-0.01500701904296875,
-0.0255126953125,
-0.007526397705078125,
0.01177215576171875,
0.0283355712890625,
-0.05255126953125,
-0.0123443603515625,
0.0192718505859375,
0.00015056133270263672,
0.058990478515625,
0.00232696533203125,
0.00867462158203125,
-0.0239410400390625,
0.041748046875,
-0.001132965087890625,
0.0546875,
0.02752685546875,
-0.03179931640625,
0.044342041015625,
0.0185546875,
-0.049102783203125,
-0.04443359375,
0.00450897216796875,
-0.0860595703125,
0.0058135986328125,
0.0877685546875,
-0.0235595703125,
-0.039947509765625,
0.0246734619140625,
-0.0226287841796875,
0.042999267578125,
-0.0240631103515625,
0.04290771484375,
0.01515960693359375,
-0.004444122314453125,
-0.00594329833984375,
-0.055908203125,
0.0282135009765625,
0.012542724609375,
-0.0689697265625,
-0.0203094482421875,
0.038330078125,
0.047119140625,
0.005687713623046875,
0.06414794921875,
-0.0037784576416015625,
0.0284423828125,
-0.0070343017578125,
0.01727294921875,
-0.0185394287109375,
-0.01236724853515625,
-0.038848876953125,
0.0072784423828125,
0.0144805908203125,
-0.0308685302734375
]
] |
akjindal53244/Mistral-7B-v0.1-Open-Platypus | 2023-10-19T02:09:14.000Z | [
"transformers",
"pytorch",
"mistral",
"text-generation",
"license:apache-2.0",
"endpoints_compatible",
"text-generation-inference",
"region:us"
] | text-generation | akjindal53244 | null | null | akjindal53244/Mistral-7B-v0.1-Open-Platypus | 3 | 7,203 | transformers | 2023-10-05T22:48:41 | ---
license: apache-2.0
---
Model is instruction-finetuned using Open-Platypus dataset: https://huggingface.co/datasets/garage-bAInd/Open-Platypus
| 149 | [
[
-0.0111083984375,
-0.053924560546875,
0.00843048095703125,
0.01898193359375,
-0.01447296142578125,
-0.03326416015625,
-0.01983642578125,
0.0026836395263671875,
-0.00545501708984375,
0.056243896484375,
-0.06646728515625,
-0.046905517578125,
-0.01177215576171875,
-0.037994384765625,
-0.031524658203125,
0.0831298828125,
-0.0005931854248046875,
0.0192718505859375,
-0.0207061767578125,
-0.0082550048828125,
-0.0276336669921875,
-0.01012420654296875,
-0.03656005859375,
-0.0213165283203125,
-0.0015869140625,
0.045562744140625,
0.05792236328125,
0.0579833984375,
0.03546142578125,
0.00749969482421875,
0.00623321533203125,
-0.006717681884765625,
-0.044097900390625,
-0.0233612060546875,
-0.003299713134765625,
-0.0272369384765625,
-0.0443115234375,
0.01947021484375,
0.04290771484375,
0.051605224609375,
-0.041107177734375,
0.03350830078125,
0.0200653076171875,
0.032684326171875,
-0.0343017578125,
0.024444580078125,
-0.01540374755859375,
-0.0096893310546875,
-0.016387939453125,
0.001529693603515625,
-0.022430419921875,
-0.01094818115234375,
-0.0010557174682617188,
-0.06536865234375,
0.037933349609375,
0.00502777099609375,
0.080810546875,
0.03314208984375,
-0.02191162109375,
-0.019561767578125,
-0.0263519287109375,
0.0307159423828125,
-0.0235748291015625,
0.034210205078125,
0.0362548828125,
0.0245361328125,
-0.02996826171875,
-0.046417236328125,
-0.01114654541015625,
-0.0157623291015625,
0.0021839141845703125,
0.005947113037109375,
0.020782470703125,
0.0142669677734375,
0.0308685302734375,
0.056549072265625,
-0.03167724609375,
0.015380859375,
-0.05841064453125,
-0.00972747802734375,
0.041717529296875,
0.01070404052734375,
0.006175994873046875,
-0.00839996337890625,
-0.03265380859375,
-0.0172576904296875,
-0.03802490234375,
0.0239715576171875,
0.029632568359375,
0.049713134765625,
-0.05694580078125,
0.06768798828125,
-0.0236053466796875,
0.064697265625,
-0.028533935546875,
0.005130767822265625,
0.039642333984375,
0.008056640625,
-0.0450439453125,
-0.00029087066650390625,
0.0380859375,
0.051055908203125,
0.0360107421875,
-0.00457000732421875,
-0.02880859375,
-0.00357818603515625,
0.015228271484375,
-0.045562744140625,
-0.080078125,
0.006923675537109375,
-0.03192138671875,
-0.004238128662109375,
0.02862548828125,
-0.044921875,
-0.00620269775390625,
-0.041717529296875,
0.05059814453125,
-0.03271484375,
-0.0173492431640625,
0.037872314453125,
-0.0002841949462890625,
0.027587890625,
0.0274658203125,
-0.04473876953125,
0.0538330078125,
0.061981201171875,
0.08648681640625,
0.0201873779296875,
-0.02166748046875,
-0.03179931640625,
0.002872467041015625,
0.00408935546875,
0.038299560546875,
-0.028228759765625,
-0.01507568359375,
-0.0270843505859375,
0.036865234375,
-0.0167999267578125,
-0.029449462890625,
0.04962158203125,
-0.02716064453125,
0.01486968994140625,
-0.0008096694946289062,
-0.04718017578125,
-0.0256195068359375,
0.005580902099609375,
-0.07086181640625,
0.05999755859375,
0.0347900390625,
-0.04205322265625,
0.01399993896484375,
-0.0860595703125,
-0.0252838134765625,
-0.007061004638671875,
0.00542449951171875,
-0.0367431640625,
0.0026721954345703125,
-0.005767822265625,
0.0306243896484375,
-0.0211181640625,
-0.006439208984375,
-0.03643798828125,
-0.001949310302734375,
0.006805419921875,
-0.00875091552734375,
0.07244873046875,
0.026580810546875,
0.0204010009765625,
0.0166473388671875,
-0.07769775390625,
-0.00939178466796875,
0.02288818359375,
-0.0007615089416503906,
-0.033416748046875,
-0.0287628173828125,
0.005298614501953125,
0.00402069091796875,
0.03533935546875,
-0.05535888671875,
0.02581787109375,
-0.00640106201171875,
0.0313720703125,
0.04583740234375,
-0.0059356689453125,
0.0264129638671875,
-0.0379638671875,
0.0401611328125,
-0.00901031494140625,
0.0268707275390625,
0.0009889602661132812,
-0.034027099609375,
-0.0491943359375,
-0.040374755859375,
0.030303955078125,
0.046356201171875,
-0.0167236328125,
0.0174713134765625,
0.02789306640625,
-0.031982421875,
-0.04180908203125,
-0.0089263916015625,
0.0235137939453125,
0.031524658203125,
0.021820068359375,
-0.0218963623046875,
-0.02728271484375,
-0.05316162109375,
0.0299530029296875,
0.0304718017578125,
-0.01338958740234375,
0.006511688232421875,
0.038482666015625,
0.00919342041015625,
0.0462646484375,
-0.059326171875,
-0.0107879638671875,
0.026885986328125,
0.00821685791015625,
0.023284912109375,
0.06671142578125,
0.041412353515625,
-0.034332275390625,
-0.029144287109375,
-0.0047149658203125,
-0.033905029296875,
-0.015655517578125,
0.01073455810546875,
-0.036224365234375,
-0.0011615753173828125,
0.0080718994140625,
-0.0726318359375,
0.04840087890625,
0.036376953125,
-0.015533447265625,
0.042388916015625,
-0.00833892822265625,
-0.0031490325927734375,
-0.06805419921875,
0.01160430908203125,
-0.00244903564453125,
-0.0201263427734375,
-0.016082763671875,
0.0272369384765625,
0.017364501953125,
0.0027904510498046875,
-0.02337646484375,
0.00589752197265625,
-0.037811279296875,
-0.022186279296875,
-0.002197265625,
-0.0069732666015625,
-0.004581451416015625,
0.0132293701171875,
0.00475311279296875,
0.05718994140625,
0.06488037109375,
-0.034027099609375,
0.026519775390625,
0.046630859375,
-0.0291595458984375,
0.01334381103515625,
-0.050079345703125,
0.007843017578125,
0.021759033203125,
-0.01629638671875,
-0.054473876953125,
-0.059326171875,
0.03338623046875,
-0.0087127685546875,
0.014312744140625,
-0.016754150390625,
-0.043609619140625,
-0.0248870849609375,
-0.0263214111328125,
0.022216796875,
0.036468505859375,
-0.053558349609375,
0.019927978515625,
0.0207672119140625,
-0.01153564453125,
-0.0129241943359375,
-0.021820068359375,
-0.040771484375,
-0.002384185791015625,
-0.037567138671875,
0.0007328987121582031,
-0.0006656646728515625,
-0.01532745361328125,
-0.01019287109375,
0.0151519775390625,
-0.018463134765625,
0.0007510185241699219,
0.0364990234375,
0.049713134765625,
-0.041656494140625,
0.008880615234375,
-0.0037078857421875,
0.01338958740234375,
0.0184783935546875,
-0.0024623870849609375,
0.064453125,
-0.0286865234375,
-0.02008056640625,
-0.048583984375,
-0.013336181640625,
0.0474853515625,
-0.01074981689453125,
0.060211181640625,
0.0263519287109375,
-0.049102783203125,
-0.0008034706115722656,
-0.0227813720703125,
-0.031585693359375,
-0.03167724609375,
0.0002300739288330078,
-0.033599853515625,
-0.0322265625,
0.05072021484375,
-0.00385284423828125,
-0.005523681640625,
0.037445068359375,
0.0184173583984375,
-0.0110015869140625,
0.06866455078125,
0.029510498046875,
0.028961181640625,
0.018768310546875,
-0.049041748046875,
-0.034027099609375,
-0.069091796875,
-0.034393310546875,
-0.0293121337890625,
-0.019744873046875,
-0.042205810546875,
-0.0200958251953125,
-0.0051116943359375,
0.035491943359375,
-0.04840087890625,
0.0272216796875,
-0.046234130859375,
0.04534912109375,
0.072509765625,
0.042205810546875,
0.00331878662109375,
-0.00860595703125,
-0.0229034423828125,
0.00287628173828125,
-0.04742431640625,
-0.051513671875,
0.08447265625,
0.032196044921875,
0.067138671875,
-0.0146026611328125,
0.03460693359375,
-0.01206207275390625,
0.02203369140625,
-0.031890869140625,
0.0272216796875,
-0.0281829833984375,
-0.03607177734375,
-0.007450103759765625,
-0.0225067138671875,
-0.048583984375,
-0.0165863037109375,
-0.01393890380859375,
-0.043426513671875,
-0.0146026611328125,
0.0240631103515625,
-0.050689697265625,
0.0234527587890625,
-0.0692138671875,
0.08905029296875,
-0.0256500244140625,
-0.0282745361328125,
-0.014190673828125,
-0.058319091796875,
0.06304931640625,
0.0092620849609375,
-0.001628875732421875,
-0.00664520263671875,
0.01214599609375,
0.05859375,
-0.06768798828125,
0.05950927734375,
-0.03582763671875,
0.00081634521484375,
0.027984619140625,
0.005405426025390625,
0.01047515869140625,
0.003299713134765625,
0.0037250518798828125,
0.006061553955078125,
0.01497650146484375,
-0.022613525390625,
0.000316619873046875,
0.05908203125,
-0.07464599609375,
0.00311279296875,
-0.0283203125,
-0.0712890625,
0.00873565673828125,
0.00510406494140625,
0.005321502685546875,
0.050079345703125,
-0.0223388671875,
0.02484130859375,
0.050384521484375,
0.00479888916015625,
0.009674072265625,
0.047698974609375,
-0.027557373046875,
-0.0309906005859375,
0.055999755859375,
0.005413055419921875,
0.01551055908203125,
0.0178680419921875,
0.02557373046875,
-0.0174713134765625,
-0.0269775390625,
-0.0247344970703125,
0.027587890625,
-0.030059814453125,
-0.03271484375,
-0.01355743408203125,
-0.033294677734375,
-0.01525115966796875,
-0.005126953125,
-0.039642333984375,
-0.053741455078125,
-0.059906005859375,
-0.01201629638671875,
0.056854248046875,
0.07086181640625,
-0.0180816650390625,
0.0655517578125,
-0.05352783203125,
0.02783203125,
0.0089263916015625,
0.058807373046875,
-0.017913818359375,
-0.040191650390625,
-0.01070404052734375,
0.0338134765625,
-0.039825439453125,
-0.032318115234375,
0.024810791015625,
0.017425537109375,
0.037750244140625,
0.0290374755859375,
0.023529052734375,
0.039581298828125,
-0.038665771484375,
0.0345458984375,
0.00551605224609375,
-0.047088623046875,
0.05584716796875,
-0.04034423828125,
0.017547607421875,
0.0299530029296875,
0.044189453125,
0.005222320556640625,
-0.0122528076171875,
-0.062164306640625,
-0.053009033203125,
0.0823974609375,
0.0159759521484375,
-0.0258026123046875,
0.0174560546875,
0.032440185546875,
0.048736572265625,
0.033905029296875,
-0.044036865234375,
0.004550933837890625,
-0.03680419921875,
-0.03741455078125,
0.0107421875,
0.01381683349609375,
-0.02288818359375,
-0.02752685546875,
0.05889892578125,
-0.001659393310546875,
0.01235198974609375,
-0.011962890625,
-0.0085906982421875,
-0.027374267578125,
-0.004367828369140625,
0.062255859375,
0.037933349609375,
-0.0655517578125,
-0.0147247314453125,
-0.02252197265625,
-0.045257568359375,
-0.0213165283203125,
0.03045654296875,
-0.0012845993041992188,
-0.03485107421875,
0.019134521484375,
0.0694580078125,
0.010345458984375,
-0.0361328125,
0.0236663818359375,
0.00365447998046875,
-0.0074005126953125,
-0.022003173828125,
0.03253173828125,
0.0011386871337890625,
0.0256500244140625,
0.00533294677734375,
0.0148773193359375,
0.0013904571533203125,
-0.01406097412109375,
0.0190277099609375,
0.0268402099609375,
-0.006351470947265625,
-0.04644775390625,
0.06494140625,
0.0141143798828125,
-0.038055419921875,
0.08416748046875,
-0.0162353515625,
-0.0248870849609375,
0.052825927734375,
0.046539306640625,
0.040283203125,
-0.0221099853515625,
0.0033969879150390625,
0.03369140625,
0.027557373046875,
-0.013946533203125,
0.06365966796875,
-0.0009508132934570312,
-0.040924072265625,
-0.0137176513671875,
-0.055267333984375,
-0.036956787109375,
0.040679931640625,
-0.090087890625,
0.038299560546875,
-0.053375244140625,
0.00640106201171875,
-0.0172119140625,
-0.0063629150390625,
-0.054443359375,
0.0257720947265625,
-0.00225067138671875,
0.10772705078125,
-0.07916259765625,
0.06298828125,
0.050689697265625,
-0.031280517578125,
-0.0655517578125,
-0.02020263671875,
-0.007137298583984375,
-0.07086181640625,
0.0209808349609375,
0.03900146484375,
0.00360107421875,
0.020355224609375,
-0.07403564453125,
-0.032745361328125,
0.099853515625,
0.051116943359375,
-0.07672119140625,
0.0099945068359375,
-0.0011444091796875,
0.015533447265625,
-0.03631591796875,
0.0004901885986328125,
0.0430908203125,
0.0161590576171875,
0.01308441162109375,
-0.08038330078125,
-0.020263671875,
-0.0180511474609375,
-0.0242767333984375,
0.025115966796875,
-0.053558349609375,
0.08758544921875,
-0.0149078369140625,
0.0004436969757080078,
0.035064697265625,
0.041412353515625,
0.01537322998046875,
0.032684326171875,
0.044830322265625,
0.07470703125,
0.05303955078125,
-0.013275146484375,
0.0615234375,
-0.039276123046875,
0.0372314453125,
0.1158447265625,
-0.02081298828125,
0.06988525390625,
0.0194244384765625,
-0.01025390625,
0.030303955078125,
0.05938720703125,
-0.00809478759765625,
0.061309814453125,
-0.013031005859375,
0.00281524658203125,
-0.0026912689208984375,
0.002193450927734375,
-0.0244598388671875,
0.03509521484375,
0.0543212890625,
-0.0260162353515625,
-0.0121002197265625,
0.0178375244140625,
-0.0270843505859375,
-0.00858306884765625,
-0.055267333984375,
0.043914794921875,
-0.006443023681640625,
-0.0213775634765625,
0.041473388671875,
-0.01019287109375,
0.0214385986328125,
-0.04290771484375,
-0.01113128662109375,
0.02996826171875,
0.0191802978515625,
-0.01123046875,
-0.08013916015625,
0.01140594482421875,
-0.0242767333984375,
-0.01515960693359375,
-0.01035308837890625,
0.053253173828125,
-0.046051025390625,
-0.031707763671875,
0.00199127197265625,
0.0113677978515625,
0.0144805908203125,
0.006244659423828125,
-0.04931640625,
0.0030059814453125,
0.0011014938354492188,
-0.004528045654296875,
-0.00879669189453125,
0.0132293701171875,
-0.00809478759765625,
0.032440185546875,
0.0191497802734375,
-0.0032196044921875,
0.01325225830078125,
0.01318359375,
0.072265625,
-0.028472900390625,
-0.040191650390625,
-0.0237884521484375,
0.044647216796875,
-0.0118408203125,
-0.042388916015625,
0.051422119140625,
0.06494140625,
0.0894775390625,
-0.043731689453125,
0.025482177734375,
-0.004974365234375,
0.043792724609375,
-0.0576171875,
0.03912353515625,
-0.0103302001953125,
-0.01517486572265625,
0.01068878173828125,
-0.048858642578125,
-0.01287841796875,
0.0748291015625,
-0.028961181640625,
0.0034351348876953125,
0.0426025390625,
0.0791015625,
-0.059814453125,
0.05279541015625,
-0.01157379150390625,
0.03338623046875,
-0.025909423828125,
0.00948333740234375,
0.059722900390625,
-0.0260162353515625,
-0.0092926025390625,
-0.0660400390625,
-0.041717529296875,
-0.013916015625,
-0.048858642578125,
-0.06842041015625,
-0.0277862548828125,
-0.04058837890625,
-0.0052337646484375,
0.00370025634765625,
0.09246826171875,
0.05291748046875,
-0.07525634765625,
-0.033233642578125,
-0.032684326171875,
-0.0272369384765625,
-0.00930023193359375,
-0.0246734619140625,
0.032440185546875,
-0.003448486328125,
-0.030059814453125,
0.005687713623046875,
-0.010345458984375,
0.036376953125,
-0.00931549072265625,
-0.01067352294921875,
-0.03167724609375,
-0.01346588134765625,
0.020538330078125,
0.013519287109375,
-0.0276336669921875,
-0.025360107421875,
-0.0174407958984375,
-0.0020656585693359375,
0.00919342041015625,
0.03717041015625,
-0.034149169921875,
0.0207061767578125,
0.015960693359375,
0.00833892822265625,
0.020843505859375,
0.01015472412109375,
0.02508544921875,
-0.0726318359375,
0.017852783203125,
-0.00803375244140625,
0.0308837890625,
0.043670654296875,
-0.02008056640625,
0.045654296875,
0.0097503662109375,
-0.050384521484375,
-0.0634765625,
-0.00936126708984375,
-0.0823974609375,
0.00966644287109375,
0.046356201171875,
-0.0087127685546875,
-0.046966552734375,
0.0239410400390625,
-0.03753662109375,
0.031768798828125,
-0.034515380859375,
0.0421142578125,
0.035552978515625,
-0.0118865966796875,
-0.019439697265625,
-0.0628662109375,
0.0099029541015625,
-0.01509857177734375,
-0.0484619140625,
0.0010776519775390625,
0.052490234375,
0.0280303955078125,
0.007781982421875,
-0.00020551681518554688,
0.01131439208984375,
0.037750244140625,
-0.00559234619140625,
0.0126800537109375,
-0.0279388427734375,
-0.050994873046875,
-0.023651123046875,
0.010223388671875,
0.0116424560546875,
-0.058319091796875
]
] |
KoboldAI/OPT-2.7B-Nerybus-Mix | 2023-02-10T05:38:20.000Z | [
"transformers",
"pytorch",
"opt",
"text-generation",
"en",
"license:other",
"has_space",
"text-generation-inference",
"region:us"
] | text-generation | KoboldAI | null | null | KoboldAI/OPT-2.7B-Nerybus-Mix | 9 | 7,198 | transformers | 2023-02-09T10:45:38 | ---
license: other
language:
- en
inference: false
---
# OPT-2.7B-Nerybus-Mix
This is an experimental model containing a ***parameter-wise 50/50 blend (weighted average)*** of the weights of *NerysV2-2.7B* and *ErebusV1-2.7B*
Preliminary testing produces pretty coherent outputs, it appears to retain the NSFWness of Erebus but with a Nerys-esque twist in terms of prose.
# License
The two models used for this blend, *NerysV2-2.7B* and *ErebusV1-2.7B* are made by **Mr. Seeker**.
- https://huggingface.co/KoboldAI/OPT-2.7B-Erebus
- https://huggingface.co/KoboldAI/OPT-2.7B-Nerys-v2
The base OPT-2.7B model is licensed under the OPT-175B license, Copyright (c) Meta Platforms, Inc. All Rights Reserved.
# Evaluation Results
As the original datasets used for the source models are not publically available, I use my own datasets for this evaluation, which may not provide accurate comparison.
Eval parameters: 32000 characters extracted from the middle of the corpus, tested in blocks of 1024 tokens each, same dataset used for each test batch.
```
Literotica Dataset Eval (Randomly selected stories)
{'eval_loss': 2.571258306503296, 'name': 'Concedo_OPT-2.7B-Nerybus-Mix'}
{'eval_loss': 2.5491442680358887, 'name': 'KoboldAI_OPT-2.7B-Erebus'}
{'eval_loss': 2.6158597469329834, 'name': 'KoboldAI_OPT-2.7B-Nerys'}
{'eval_loss': 2.614469051361084, 'name': 'facebook_opt-2.7b'}
{'eval_loss': 2.4960227012634277, 'name': '(Unreleased 2.7B ModronAI Model)'}
ASSTR Dataset Eval (Randomly selected stories)
{'eval_loss': 2.664412498474121, 'name': 'Concedo_OPT-2.7B-Nerybus-Mix'}
{'eval_loss': 2.6451029777526855, 'name': 'KoboldAI_OPT-2.7B-Erebus'}
{'eval_loss': 2.7259647846221924, 'name': 'KoboldAI_OPT-2.7B-Nerys'}
{'eval_loss': 2.6675195693969727, 'name': 'facebook_opt-2.7b'}
{'eval_loss': 2.962111473083496, 'name': '(Unreleased 2.7B ModronAI Model)'}
Sexstories Dataset Eval (Random highly rated stories)
{'eval_loss': 2.2352423667907715, 'name': 'Concedo_OPT-2.7B-Nerybus-Mix'}
{'eval_loss': 2.194378137588501, 'name': 'KoboldAI_OPT-2.7B-Erebus'}
{'eval_loss': 2.307469129562378, 'name': 'KoboldAI_OPT-2.7B-Nerys'}
{'eval_loss': 2.293961763381958, 'name': 'facebook_opt-2.7b'}
{'eval_loss': 2.0103421211242676, 'name': '(Unreleased 2.7B ModronAI Model)'}
Harry Potter Dataset Eval (Canon books)
{'eval_loss': 2.473742961883545, 'name': 'Concedo_OPT-2.7B-Nerybus-Mix'}
{'eval_loss': 2.480600357055664, 'name': 'KoboldAI_OPT-2.7B-Erebus'}
{'eval_loss': 2.506237506866455, 'name': 'KoboldAI_OPT-2.7B-Nerys'}
{'eval_loss': 2.5074169635772705, 'name': 'facebook_opt-2.7b'}
{'eval_loss': 2.273703098297119, 'name': '(Unreleased 2.7B ModronAI Model)'}
Star Wars Dataset Eval (Rogue One Novel)
{'eval_loss': 2.5031676292419434, 'name': 'Concedo_OPT-2.7B-Nerybus-Mix'}
{'eval_loss': 2.5239150524139404, 'name': 'KoboldAI_OPT-2.7B-Erebus'}
{'eval_loss': 2.526801586151123, 'name': 'KoboldAI_OPT-2.7B-Nerys'}
{'eval_loss': 2.473283529281616, 'name': 'facebook_opt-2.7b'}
{'eval_loss': 2.955465793609619, 'name': '(Unreleased 2.7B ModronAI Model)'}
```
It is recommend to use this model with the KoboldAI software. All feedback and comments can be directed to Concedo on the KoboldAI discord.
| 3,199 | [
[
-0.03076171875,
-0.05267333984375,
0.005886077880859375,
0.0211639404296875,
-0.0124664306640625,
-0.01221466064453125,
-0.01453399658203125,
-0.0345458984375,
0.050201416015625,
0.03485107421875,
-0.058563232421875,
-0.055999755859375,
-0.035400390625,
0.0018053054809570312,
-0.0255279541015625,
0.07110595703125,
0.0035991668701171875,
0.01088714599609375,
0.00848388671875,
-0.00937652587890625,
-0.0294647216796875,
-0.030792236328125,
-0.05072021484375,
-0.013824462890625,
0.047760009765625,
0.028900146484375,
0.061126708984375,
0.0248870849609375,
0.037384033203125,
0.0182952880859375,
-0.02874755859375,
0.01309967041015625,
-0.052001953125,
-0.00720977783203125,
0.0091552734375,
-0.053466796875,
-0.04779052734375,
0.010986328125,
0.04541015625,
0.039520263671875,
-0.006496429443359375,
0.024444580078125,
-0.0083465576171875,
0.047149658203125,
-0.024200439453125,
-0.01290130615234375,
-0.015472412109375,
0.007343292236328125,
-0.018035888671875,
-0.00025081634521484375,
-0.034332275390625,
-0.02764892578125,
-0.0012922286987304688,
-0.05487060546875,
0.017364501953125,
0.041748046875,
0.1019287109375,
-0.0016117095947265625,
-0.009063720703125,
-0.025634765625,
-0.027618408203125,
0.0762939453125,
-0.0880126953125,
0.01806640625,
0.021514892578125,
0.0058441162109375,
-0.019744873046875,
-0.0300140380859375,
-0.045013427734375,
-0.0032253265380859375,
-0.01464080810546875,
0.0226898193359375,
-0.02685546875,
-0.0234527587890625,
0.0128326416015625,
0.03546142578125,
-0.061126708984375,
0.0136260986328125,
-0.037750244140625,
-0.00821685791015625,
0.06005859375,
0.02642822265625,
0.0196380615234375,
-0.0296783447265625,
-0.0279693603515625,
-0.0303192138671875,
-0.03509521484375,
0.01447296142578125,
0.0635986328125,
0.0247802734375,
-0.028045654296875,
0.052947998046875,
-0.01242828369140625,
0.05810546875,
0.01247406005859375,
-0.00875091552734375,
0.055938720703125,
-0.040313720703125,
-0.0287017822265625,
0.0005068778991699219,
0.07073974609375,
0.045135498046875,
0.01009368896484375,
0.025482177734375,
0.0019445419311523438,
-0.0069580078125,
-0.0025787353515625,
-0.07403564453125,
-0.0177001953125,
0.02667236328125,
-0.055938720703125,
-0.036834716796875,
0.017669677734375,
-0.0723876953125,
-0.02337646484375,
-0.004932403564453125,
0.029327392578125,
-0.03741455078125,
-0.0288543701171875,
0.0036411285400390625,
-0.0274200439453125,
0.02783203125,
0.01177215576171875,
-0.0528564453125,
0.022796630859375,
0.0187530517578125,
0.0501708984375,
-0.008056640625,
-0.0095977783203125,
0.0038299560546875,
-0.010101318359375,
-0.036224365234375,
0.045013427734375,
-0.0216827392578125,
-0.031707763671875,
-0.0245208740234375,
0.015960693359375,
-0.01316070556640625,
-0.0292510986328125,
0.050811767578125,
-0.005039215087890625,
0.0191802978515625,
-0.0221099853515625,
-0.04913330078125,
-0.030059814453125,
0.009613037109375,
-0.056304931640625,
0.08184814453125,
0.0212249755859375,
-0.057159423828125,
0.04168701171875,
-0.0176544189453125,
-0.0010461807250976562,
-0.00925445556640625,
0.005512237548828125,
-0.044891357421875,
-0.0013580322265625,
0.026336669921875,
0.033843994140625,
-0.0290985107421875,
0.00396728515625,
-0.04864501953125,
-0.0282135009765625,
0.03399658203125,
-0.0309600830078125,
0.0875244140625,
0.010101318359375,
-0.01427459716796875,
0.00177764892578125,
-0.06475830078125,
0.0182342529296875,
0.025299072265625,
-0.0037708282470703125,
-0.0208740234375,
-0.0195159912109375,
0.0224609375,
0.007354736328125,
0.00618743896484375,
-0.030914306640625,
0.0152587890625,
-0.04522705078125,
0.01739501953125,
0.0455322265625,
0.009368896484375,
0.0209808349609375,
-0.034820556640625,
0.043548583984375,
0.005008697509765625,
0.027099609375,
0.009552001953125,
-0.043731689453125,
-0.048553466796875,
-0.035675048828125,
0.04071044921875,
0.0279998779296875,
-0.02374267578125,
0.037750244140625,
-0.01280975341796875,
-0.07427978515625,
-0.04986572265625,
-0.005634307861328125,
0.0211639404296875,
0.032196044921875,
0.01299285888671875,
-0.0006804466247558594,
-0.039215087890625,
-0.08612060546875,
-0.0135498046875,
-0.01259613037109375,
0.0025005340576171875,
0.041168212890625,
0.04046630859375,
-0.00568389892578125,
0.07269287109375,
-0.051239013671875,
-0.01739501953125,
-0.030914306640625,
0.01052093505859375,
0.056884765625,
0.03985595703125,
0.053375244140625,
-0.07373046875,
-0.048095703125,
0.00498199462890625,
-0.05621337890625,
-0.0169677734375,
-0.0021762847900390625,
-0.0201873779296875,
0.019805908203125,
0.0173187255859375,
-0.05517578125,
0.0467529296875,
0.0347900390625,
-0.053131103515625,
0.0462646484375,
-0.0219268798828125,
0.0285491943359375,
-0.08917236328125,
0.006633758544921875,
-0.0006551742553710938,
-0.00864410400390625,
-0.056396484375,
-0.016357421875,
0.0141754150390625,
0.021026611328125,
-0.0498046875,
0.066162109375,
-0.041351318359375,
-0.0019855499267578125,
0.012237548828125,
0.0174102783203125,
0.006511688232421875,
0.03985595703125,
-0.0111083984375,
0.03662109375,
0.04107666015625,
-0.019500732421875,
0.0253143310546875,
0.026153564453125,
-0.0196990966796875,
0.040191650390625,
-0.04620361328125,
-0.02459716796875,
-0.002979278564453125,
0.011962890625,
-0.0772705078125,
-0.0200347900390625,
0.0311279296875,
-0.052947998046875,
0.01277923583984375,
-0.0013837814331054688,
-0.0286407470703125,
-0.046173095703125,
-0.04241943359375,
0.019195556640625,
0.041259765625,
-0.0219573974609375,
0.049163818359375,
0.01435089111328125,
-0.016815185546875,
-0.05072021484375,
-0.06884765625,
-0.0154266357421875,
-0.0252685546875,
-0.057281494140625,
0.0294647216796875,
-0.0254669189453125,
-0.00827789306640625,
0.0184478759765625,
-0.00629425048828125,
-0.003753662109375,
-0.026336669921875,
0.0230560302734375,
0.04931640625,
0.0008950233459472656,
-0.01384735107421875,
0.0131378173828125,
-0.01119232177734375,
-0.0040740966796875,
0.00443267822265625,
0.04681396484375,
-0.006885528564453125,
-0.0101776123046875,
-0.029754638671875,
0.0073394775390625,
0.03546142578125,
-0.00238037109375,
0.06170654296875,
0.049591064453125,
-0.0281982421875,
-0.0099029541015625,
-0.029541015625,
0.0023136138916015625,
-0.031646728515625,
0.035736083984375,
-0.02197265625,
-0.041259765625,
0.07965087890625,
0.0321044921875,
-0.0030059814453125,
0.07049560546875,
0.0538330078125,
0.002979278564453125,
0.07666015625,
0.017547607421875,
-0.003082275390625,
0.0262451171875,
-0.0450439453125,
0.0195465087890625,
-0.054595947265625,
-0.038726806640625,
-0.02099609375,
-0.045440673828125,
-0.0482177734375,
-0.019989013671875,
0.01285552978515625,
0.0175018310546875,
-0.009613037109375,
0.0472412109375,
-0.040191650390625,
0.0252685546875,
0.05096435546875,
0.026641845703125,
0.003505706787109375,
-0.006855010986328125,
0.001983642578125,
-0.01068115234375,
-0.056121826171875,
0.00152587890625,
0.1005859375,
0.018768310546875,
0.060211181640625,
0.0037708282470703125,
0.051361083984375,
0.0155792236328125,
0.01751708984375,
-0.03369140625,
0.04559326171875,
-0.005035400390625,
-0.07379150390625,
-0.019744873046875,
-0.0474853515625,
-0.05029296875,
0.025238037109375,
-0.034576416015625,
-0.053558349609375,
0.047149658203125,
-0.00318145751953125,
-0.04296875,
0.01316070556640625,
-0.054962158203125,
0.07061767578125,
-0.0107421875,
-0.03399658203125,
0.0012874603271484375,
-0.051666259765625,
0.018310546875,
0.004749298095703125,
0.016265869140625,
-0.0145263671875,
0.00004202127456665039,
0.07135009765625,
-0.035430908203125,
0.04620361328125,
-0.00130462646484375,
0.0080413818359375,
0.036285400390625,
-0.00543975830078125,
0.0377197265625,
0.01457977294921875,
0.006374359130859375,
0.026641845703125,
0.002178192138671875,
-0.026885986328125,
-0.0279541015625,
0.058013916015625,
-0.048828125,
-0.00685882568359375,
-0.03326416015625,
-0.055816650390625,
0.02032470703125,
0.0274810791015625,
0.046417236328125,
0.047698974609375,
-0.006198883056640625,
0.0268402099609375,
0.04522705078125,
-0.015533447265625,
0.0380859375,
0.033203125,
-0.02703857421875,
-0.0574951171875,
0.0635986328125,
0.0157928466796875,
0.0210418701171875,
0.00677490234375,
0.005092620849609375,
-0.0286407470703125,
-0.038238525390625,
-0.0271148681640625,
0.02685546875,
-0.040313720703125,
-0.0252685546875,
-0.04071044921875,
-0.0243072509765625,
-0.026153564453125,
-0.0263824462890625,
-0.0279693603515625,
-0.0322265625,
-0.040435791015625,
-0.0150909423828125,
0.029693603515625,
0.043121337890625,
0.0012235641479492188,
0.018951416015625,
-0.035186767578125,
0.01262664794921875,
-0.002597808837890625,
0.01812744140625,
-0.0150604248046875,
-0.055908203125,
-0.01009368896484375,
0.01629638671875,
-0.004749298095703125,
-0.089111328125,
0.07354736328125,
-0.004619598388671875,
0.0257415771484375,
0.0304107666015625,
-0.00849151611328125,
0.064208984375,
-0.0017938613891601562,
0.06317138671875,
0.01678466796875,
-0.043731689453125,
0.026519775390625,
-0.03802490234375,
0.00420379638671875,
0.051605224609375,
0.0175323486328125,
-0.03131103515625,
-0.0435791015625,
-0.06695556640625,
-0.08056640625,
0.08416748046875,
0.04803466796875,
-0.00913238525390625,
0.0177154541015625,
0.01016998291015625,
0.005008697509765625,
0.02825927734375,
-0.07427978515625,
-0.050750732421875,
0.003261566162109375,
-0.03997802734375,
-0.0039825439453125,
-0.02264404296875,
-0.000835418701171875,
-0.0386962890625,
0.068115234375,
0.0205841064453125,
0.011383056640625,
0.0222625732421875,
0.0022449493408203125,
-0.006175994873046875,
-0.0008749961853027344,
0.035980224609375,
0.044708251953125,
-0.043060302734375,
-0.0232086181640625,
0.01324462890625,
-0.051116943359375,
-0.011077880859375,
0.032928466796875,
-0.0301055908203125,
0.005741119384765625,
0.019256591796875,
0.06317138671875,
0.0278472900390625,
-0.05230712890625,
0.0465087890625,
-0.00017631053924560547,
-0.039886474609375,
-0.033905029296875,
0.0025196075439453125,
0.00933074951171875,
0.033233642578125,
0.0235595703125,
0.010101318359375,
0.0090179443359375,
-0.0267333984375,
0.0028820037841796875,
0.0205535888671875,
-0.0274658203125,
-0.009857177734375,
0.056121826171875,
-0.0170440673828125,
-0.01409912109375,
0.03656005859375,
-0.0272216796875,
-0.00914764404296875,
0.0770263671875,
0.033233642578125,
0.0621337890625,
-0.0181732177734375,
0.019317626953125,
0.044189453125,
0.0301666259765625,
-0.0143890380859375,
0.03668212890625,
0.050537109375,
-0.0380859375,
0.0013933181762695312,
-0.056396484375,
0.00002187490463256836,
0.014739990234375,
-0.04498291015625,
0.045440673828125,
-0.00922393798828125,
-0.01788330078125,
0.006496429443359375,
0.00708770751953125,
-0.033416748046875,
0.0160369873046875,
0.00478363037109375,
0.056640625,
-0.084228515625,
0.0293731689453125,
0.04473876953125,
-0.041351318359375,
-0.08355712890625,
-0.03143310546875,
-0.013397216796875,
-0.031585693359375,
0.02593994140625,
0.013763427734375,
0.045013427734375,
-0.0020008087158203125,
-0.044189453125,
-0.08233642578125,
0.103759765625,
0.0172271728515625,
-0.038299560546875,
0.00024390220642089844,
0.001132965087890625,
0.040924072265625,
-0.031707763671875,
0.055084228515625,
0.031524658203125,
0.04248046875,
0.00848388671875,
-0.04962158203125,
-0.0026187896728515625,
-0.034576416015625,
-0.0076904296875,
0.0181427001953125,
-0.06005859375,
0.0943603515625,
-0.0030269622802734375,
0.00629425048828125,
0.004974365234375,
0.039794921875,
0.0308990478515625,
0.0389404296875,
0.0206756591796875,
0.064208984375,
0.05029296875,
-0.01500701904296875,
0.07861328125,
-0.0235595703125,
0.057373046875,
0.061126708984375,
0.01020050048828125,
0.04608154296875,
0.01751708984375,
-0.0263824462890625,
0.036895751953125,
0.0654296875,
-0.0177154541015625,
0.0293121337890625,
-0.00970458984375,
0.01255035400390625,
-0.00982666015625,
0.0144195556640625,
-0.02410888671875,
0.01120758056640625,
0.016815185546875,
-0.039337158203125,
-0.0007104873657226562,
-0.00994873046875,
0.027069091796875,
-0.0111083984375,
-0.0288848876953125,
0.0244903564453125,
0.0013427734375,
-0.0274810791015625,
0.0457763671875,
-0.006500244140625,
0.045166015625,
-0.058013916015625,
0.0015506744384765625,
-0.01041412353515625,
0.0100250244140625,
-0.0249481201171875,
-0.050689697265625,
0.0032749176025390625,
-0.00801849365234375,
-0.0198516845703125,
0.00478363037109375,
0.054779052734375,
-0.04400634765625,
-0.0772705078125,
0.01568603515625,
0.018951416015625,
0.011871337890625,
0.0003342628479003906,
-0.05560302734375,
0.0013322830200195312,
0.0169677734375,
-0.054290771484375,
-0.0022411346435546875,
0.0144500732421875,
0.004947662353515625,
0.049163818359375,
0.051361083984375,
0.01253509521484375,
0.01360321044921875,
-0.0088653564453125,
0.06591796875,
-0.048309326171875,
-0.047760009765625,
-0.058319091796875,
0.040679931640625,
-0.0281982421875,
-0.051422119140625,
0.08013916015625,
0.059051513671875,
0.04046630859375,
-0.0026149749755859375,
0.031768798828125,
-0.04034423828125,
0.03692626953125,
-0.0301666259765625,
0.060211181640625,
-0.06097412109375,
-0.0076904296875,
-0.0253143310546875,
-0.0791015625,
-0.007305145263671875,
0.0313720703125,
-0.0228271484375,
0.01690673828125,
0.055084228515625,
0.07525634765625,
0.004863739013671875,
0.0038890838623046875,
0.0031528472900390625,
0.039093017578125,
0.0007195472717285156,
0.038726806640625,
0.02276611328125,
-0.048919677734375,
0.0380859375,
-0.0264739990234375,
-0.0200653076171875,
-0.032379150390625,
-0.0438232421875,
-0.0687255859375,
-0.0233154296875,
-0.0050811767578125,
-0.0299072265625,
0.0022792816162109375,
0.05450439453125,
0.046112060546875,
-0.054962158203125,
-0.01206207275390625,
-0.00775146484375,
-0.0026092529296875,
-0.0273895263671875,
-0.0232696533203125,
0.028106689453125,
0.0191802978515625,
-0.06781005859375,
0.01065826416015625,
0.001251220703125,
0.0099945068359375,
-0.00463104248046875,
-0.0171661376953125,
-0.0215301513671875,
0.01342010498046875,
0.021881103515625,
0.002288818359375,
-0.043731689453125,
0.004344940185546875,
0.01107025146484375,
0.0011396408081054688,
-0.00025653839111328125,
0.00925445556640625,
-0.04278564453125,
0.02392578125,
0.055999755859375,
0.003757476806640625,
0.049591064453125,
0.01113128662109375,
0.007175445556640625,
-0.044219970703125,
0.024658203125,
0.01462554931640625,
0.022430419921875,
0.0124664306640625,
-0.037506103515625,
0.0389404296875,
0.0275421142578125,
-0.03546142578125,
-0.07843017578125,
-0.0217132568359375,
-0.10137939453125,
-0.0124969482421875,
0.0972900390625,
0.0142974853515625,
-0.0458984375,
0.031951904296875,
-0.021026611328125,
0.0018091201782226562,
-0.042877197265625,
0.01947021484375,
0.055694580078125,
-0.004161834716796875,
0.003742218017578125,
-0.057281494140625,
0.0169219970703125,
0.0281219482421875,
-0.0479736328125,
0.007183074951171875,
0.028900146484375,
0.0204010009765625,
0.00714874267578125,
0.050750732421875,
-0.0218353271484375,
0.00701141357421875,
0.0247344970703125,
0.019866943359375,
-0.0163116455078125,
-0.0176544189453125,
-0.01324462890625,
-0.0011987686157226562,
-0.036163330078125,
0.0004892349243164062
]
] |
TheBloke/Lemur-70B-Chat-v1-GPTQ | 2023-09-27T12:46:36.000Z | [
"transformers",
"safetensors",
"llama",
"text-generation",
"code",
"text-generation-inference",
"en",
"license:cc-by-nc-4.0",
"has_space",
"region:us"
] | text-generation | TheBloke | null | null | TheBloke/Lemur-70B-Chat-v1-GPTQ | 3 | 7,196 | transformers | 2023-08-29T14:02:10 | ---
language:
- en
license: cc-by-nc-4.0
library_name: transformers
tags:
- text-generation
- code
- text-generation-inference
model_name: Lemur 70B Chat v1
base_model: OpenLemur/lemur-70b-chat-v1
inference: false
model_creator: OpenLemur
model_type: llama
pipeline_tag: text-generation
prompt_template: '<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'
quantized_by: TheBloke
widget:
- example_title: Lemur favorite fruit
group: Python
text: What's lemur's favorite fruit?
- example_title: Merge Sort
group: Python
text: Write a Python function to merge two sorted lists into one sorted list without
using any built-in sort functions.
---
<!-- 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 -->
# Lemur 70B Chat v1 - GPTQ
- Model creator: [OpenLemur](https://huggingface.co/OpenLemur)
- Original model: [Lemur 70B Chat v1](https://huggingface.co/OpenLemur/lemur-70b-chat-v1)
<!-- description start -->
## Description
This repo contains GPTQ model files for [OpenLemur's Lemur 70B Chat v1](https://huggingface.co/OpenLemur/lemur-70b-chat-v1).
Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
<!-- description end -->
<!-- repositories-available start -->
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GGUF)
* [OpenLemur's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/OpenLemur/lemur-70b-chat-v1)
<!-- repositories-available end -->
<!-- prompt-template start -->
## Prompt template: ChatML
```
<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
```
<!-- prompt-template end -->
<!-- licensing start -->
## Licensing
The creator of the source model has listed its license as `cc-by-nc-4.0`, and this quantization has therefore used that same license.
As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.
In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: [OpenLemur's Lemur 70B Chat v1](https://huggingface.co/OpenLemur/lemur-70b-chat-v1).
<!-- licensing end -->
<!-- README_GPTQ.md-provided-files start -->
## Provided files and GPTQ parameters
Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.
Each separate quant is in a different branch. See below for instructions on fetching from different branches.
All recent GPTQ files are made with AutoGPTQ, and all files in non-main branches are made with AutoGPTQ. Files in the `main` branch which were uploaded before August 2023 were made with GPTQ-for-LLaMa.
<details>
<summary>Explanation of GPTQ parameters</summary>
- Bits: The bit size of the quantised model.
- GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
- Act Order: True or False. Also known as `desc_act`. True results in better quantisation accuracy. Some GPTQ clients have had issues with models that use Act Order plus Group Size, but this is generally resolved now.
- Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
- GPTQ dataset: The dataset used for quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
- Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
- ExLlama Compatibility: Whether this file can be loaded with ExLlama, which currently only supports Llama models in 4-bit.
</details>
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
| [main](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ/tree/main) | 4 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 35.33 GB | Yes | 4-bit, with Act Order. No group size, to lower VRAM requirements. |
| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 40.66 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
| [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 37.99 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
| [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 36.65 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
| [gptq-3bit--1g-actorder_True](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ/tree/gptq-3bit--1g-actorder_True) | 3 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 26.78 GB | No | 3-bit, with Act Order and no group size. Lowest possible VRAM requirements. May be lower quality than 3-bit 128g. |
| [gptq-3bit-128g-actorder_True](https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ/tree/gptq-3bit-128g-actorder_True) | 3 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 4096 | 28.03 GB | No | 3-bit, with group size 128g and act-order. Higher quality than 128g-False. |
<!-- README_GPTQ.md-provided-files end -->
<!-- README_GPTQ.md-download-from-branches start -->
## How to download from branches
- In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/Lemur-70B-Chat-v1-GPTQ:main`
- With Git, you can clone a branch with:
```
git clone --single-branch --branch main https://huggingface.co/TheBloke/Lemur-70B-Chat-v1-GPTQ
```
- In Python Transformers code, the branch is the `revision` parameter; see below.
<!-- README_GPTQ.md-download-from-branches end -->
<!-- README_GPTQ.md-text-generation-webui start -->
## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.
1. Click the **Model tab**.
2. Under **Download custom model or LoRA**, enter `TheBloke/Lemur-70B-Chat-v1-GPTQ`.
- To download from a specific branch, enter for example `TheBloke/Lemur-70B-Chat-v1-GPTQ:main`
- see Provided Files above for the list of branches for each option.
3. Click **Download**.
4. The model will start downloading. Once it's finished it will say "Done".
5. In the top left, click the refresh icon next to **Model**.
6. In the **Model** dropdown, choose the model you just downloaded: `Lemur-70B-Chat-v1-GPTQ`
7. The model will automatically load, and is now ready for use!
8. If you want any custom settings, set them and then click **Save settings for this model** followed by **Reload the Model** in the top right.
* Note that you do not need to and should not set manual GPTQ parameters any more. These are set automatically from the file `quantize_config.json`.
9. Once you're ready, click the **Text Generation tab** and enter a prompt to get started!
<!-- README_GPTQ.md-text-generation-webui end -->
<!-- README_GPTQ.md-use-from-python start -->
## How to use this GPTQ model from Python code
### Install the necessary packages
Requires: Transformers 4.32.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.
```shell
pip3 install transformers>=4.32.0 optimum>=1.12.0
pip3 install auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/ # Use cu117 if on CUDA 11.7
```
If you have problems installing AutoGPTQ using the pre-built wheels, install it from source instead:
```shell
pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
pip3 install .
```
### For CodeLlama models only: you must use Transformers 4.33.0 or later.
If 4.33.0 is not yet released when you read this, you will need to install Transformers from source:
```shell
pip3 uninstall -y transformers
pip3 install git+https://github.com/huggingface/transformers.git
```
### You can then use the following code
```python
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
model_name_or_path = "TheBloke/Lemur-70B-Chat-v1-GPTQ"
# To use a different branch, change revision
# For example: revision="main"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
device_map="auto",
trust_remote_code=False,
revision="main")
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
prompt = "Tell me about AI"
prompt_template=f'''<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
'''
print("\n\n*** Generate:")
input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
print(tokenizer.decode(output[0]))
# Inference can also be done using transformers' pipeline
print("*** Pipeline:")
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
max_new_tokens=512,
do_sample=True,
temperature=0.7,
top_p=0.95,
top_k=40,
repetition_penalty=1.1
)
print(pipe(prompt_template)[0]['generated_text'])
```
<!-- README_GPTQ.md-use-from-python end -->
<!-- README_GPTQ.md-compatibility start -->
## Compatibility
The files provided are tested to work with AutoGPTQ, both via Transformers and using AutoGPTQ directly. They should also work with [Occ4m's GPTQ-for-LLaMa fork](https://github.com/0cc4m/KoboldAI).
[ExLlama](https://github.com/turboderp/exllama) is compatible with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
[Huggingface Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) is compatible with all GPTQ models.
<!-- README_GPTQ.md-compatibility end -->
<!-- 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!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
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**: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end -->
# Original model card: OpenLemur's Lemur 70B Chat v1
# lemur-70b-chat-v1
<p align="center">
<img src="https://huggingface.co/datasets/OpenLemur/assets/resolve/main/lemur_icon.png" width="300" height="300" alt="Lemur">
</p>
<div align="center">
<img src="https://huggingface.co/datasets/OpenLemur/assets/resolve/main/lemur_chat_radar.png">
</div>
## Use
### Setup
First, we have to install all the libraries listed in `requirements.txt` in [GitHub](https://github.com/OpenLemur/lemur-v1):
```bash
pip install -r requirements.txt
```
### Generation
```python
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("OpenLemur/lemur-70b-chat-v1")
model = AutoModelForCausalLM.from_pretrained("OpenLemur/lemur-70b-chat-v1", device_map="auto", load_in_8bit=True)
# Text Generation Example
prompt = """<|im_start|>system
You are a helpful, respectful, and honest assistant.
<|im_end|>
<|im_start|>user
What's a lemur's favorite fruit?<|im_end|>
<|im_start|>assistant
"""
input = tokenizer(prompt, return_tensors="pt")
output = model.generate(**input, max_length=50, num_return_sequences=1)
generated_text = tokenizer.decode(output[0], skip_special_tokens=True)
print(generated_text)
# Code Generation Example
prompt = """<|im_start|>system
Below is an instruction that describes a task. Write a response that appropriately completes the request.
<|im_end|>
<|im_start|>user
Write a Python function to merge two sorted lists into one sorted list without using any built-in sort functions.<|im_end|>
<|im_start|>assistant
"""
input = tokenizer(prompt, return_tensors="pt")
output = model.generate(**input, max_length=200, num_return_sequences=1)
generated_code = tokenizer.decode(output[0], skip_special_tokens=True)
print(generated_code)
```
# License
The model is licensed under a CC BY-NC-4.0 license focused on research use cases.
# Acknowledgements
The Lemur project is an open collaborative research effort between [XLang Lab](https://www.xlang.ai/) and Salesforce Research. We thank Salesforce, Google Research and Amazon AWS for their gift support.
| 18,071 | [
[
-0.0404052734375,
-0.0604248046875,
-0.0041656494140625,
0.032440185546875,
-0.022735595703125,
-0.00567626953125,
0.001556396484375,
-0.046112060546875,
0.0207366943359375,
0.026123046875,
-0.04339599609375,
-0.032684326171875,
-0.0254669189453125,
-0.0087127685546875,
-0.016021728515625,
0.08355712890625,
0.0166473388671875,
-0.02447509765625,
-0.01259613037109375,
-0.0294342041015625,
-0.0184478759765625,
-0.0452880859375,
-0.061553955078125,
-0.01457977294921875,
0.02978515625,
0.02142333984375,
0.0670166015625,
0.03228759765625,
0.0184478759765625,
0.02239990234375,
-0.01447296142578125,
-0.0026645660400390625,
-0.048858642578125,
-0.017852783203125,
0.0120086669921875,
-0.01137542724609375,
-0.0570068359375,
0.0049591064453125,
0.03155517578125,
0.0195770263671875,
-0.0274200439453125,
0.01690673828125,
0.007579803466796875,
0.0494384765625,
-0.035369873046875,
0.0158233642578125,
-0.0272674560546875,
-0.002777099609375,
-0.01120758056640625,
0.0100250244140625,
-0.01088714599609375,
-0.0259552001953125,
0.0068206787109375,
-0.056365966796875,
-0.00926971435546875,
0.0006093978881835938,
0.0831298828125,
0.0012159347534179688,
-0.042083740234375,
0.00616455078125,
-0.0274810791015625,
0.047454833984375,
-0.07562255859375,
0.0268402099609375,
0.044281005859375,
0.0193023681640625,
-0.0179595947265625,
-0.058807373046875,
-0.039459228515625,
-0.0000947117805480957,
-0.01544952392578125,
0.00992584228515625,
-0.034210205078125,
0.0006365776062011719,
0.0293121337890625,
0.05084228515625,
-0.06378173828125,
-0.00829315185546875,
-0.0305633544921875,
-0.0211029052734375,
0.0638427734375,
0.005420684814453125,
0.0286407470703125,
-0.01425933837890625,
-0.0166473388671875,
-0.02978515625,
-0.046630859375,
0.0113525390625,
0.028350830078125,
0.008148193359375,
-0.056488037109375,
0.04144287109375,
-0.0302734375,
0.034820556640625,
0.0087738037109375,
-0.016357421875,
0.032989501953125,
-0.04534912109375,
-0.034027099609375,
-0.025360107421875,
0.0982666015625,
0.03753662109375,
-0.011383056640625,
0.02349853515625,
0.00768280029296875,
-0.015533447265625,
-0.0132904052734375,
-0.07293701171875,
-0.038909912109375,
0.0299224853515625,
-0.03753662109375,
-0.02288818359375,
-0.01263427734375,
-0.054046630859375,
0.001689910888671875,
0.00653839111328125,
0.031280517578125,
-0.047821044921875,
-0.04376220703125,
0.006561279296875,
-0.0171661376953125,
0.034912109375,
0.024566650390625,
-0.059326171875,
0.044219970703125,
0.02386474609375,
0.062347412109375,
0.01496124267578125,
-0.005107879638671875,
-0.0203704833984375,
0.0014085769653320312,
-0.0033588409423828125,
0.0311431884765625,
-0.0185546875,
-0.036895751953125,
-0.0307769775390625,
0.019073486328125,
0.00768280029296875,
-0.01617431640625,
0.0280914306640625,
-0.01995849609375,
0.032073974609375,
-0.03619384765625,
-0.03009033203125,
-0.0222320556640625,
0.0074920654296875,
-0.0457763671875,
0.09521484375,
0.0296630859375,
-0.0732421875,
0.013824462890625,
-0.046722412109375,
-0.01483154296875,
-0.00759124755859375,
-0.0008878707885742188,
-0.038299560546875,
-0.0194854736328125,
0.0218048095703125,
0.023590087890625,
-0.021484375,
0.0003044605255126953,
-0.036041259765625,
-0.0170135498046875,
0.01739501953125,
-0.02899169921875,
0.10064697265625,
0.00856781005859375,
-0.02215576171875,
-0.00585174560546875,
-0.041412353515625,
0.006282806396484375,
0.0294647216796875,
-0.01641845703125,
-0.0108489990234375,
-0.00673675537109375,
0.0013055801391601562,
0.0161285400390625,
0.01629638671875,
-0.031768798828125,
0.03289794921875,
-0.0207366943359375,
0.045867919921875,
0.03973388671875,
0.005031585693359375,
0.023406982421875,
-0.038726806640625,
0.0284576416015625,
0.0081634521484375,
0.0528564453125,
0.00750732421875,
-0.056060791015625,
-0.04632568359375,
-0.0310211181640625,
0.02642822265625,
0.0472412109375,
-0.038421630859375,
0.034210205078125,
-0.00972747802734375,
-0.0511474609375,
-0.03167724609375,
-0.006290435791015625,
0.0213165283203125,
0.01181793212890625,
0.029388427734375,
-0.039825439453125,
-0.033935546875,
-0.06549072265625,
0.01404571533203125,
-0.038665771484375,
0.006908416748046875,
0.052490234375,
0.05303955078125,
-0.0163726806640625,
0.054779052734375,
-0.03314208984375,
-0.006252288818359375,
-0.00011962652206420898,
0.003631591796875,
0.0209808349609375,
0.048675537109375,
0.07025146484375,
-0.060394287109375,
-0.051116943359375,
-0.004138946533203125,
-0.048797607421875,
-0.0038928985595703125,
0.00036597251892089844,
-0.034027099609375,
0.0240478515625,
0.0024929046630859375,
-0.08221435546875,
0.054046630859375,
0.049102783203125,
-0.03582763671875,
0.047882080078125,
-0.005580902099609375,
0.01465606689453125,
-0.07574462890625,
0.00632476806640625,
-0.004459381103515625,
-0.027069091796875,
-0.02972412109375,
-0.004962921142578125,
-0.0114898681640625,
0.0185546875,
-0.033203125,
0.052459716796875,
-0.040618896484375,
-0.00975799560546875,
0.0108795166015625,
0.0046539306640625,
0.0111083984375,
0.039703369140625,
-0.01381683349609375,
0.054046630859375,
0.045166015625,
-0.030548095703125,
0.045074462890625,
0.03900146484375,
0.0007839202880859375,
0.0281524658203125,
-0.06805419921875,
0.0211639404296875,
0.017181396484375,
0.031280517578125,
-0.07464599609375,
-0.0210723876953125,
0.051239013671875,
-0.052276611328125,
0.03009033203125,
-0.024200439453125,
-0.035797119140625,
-0.032623291015625,
-0.044464111328125,
0.0291595458984375,
0.060394287109375,
-0.02337646484375,
0.0248565673828125,
0.036834716796875,
-0.0014657974243164062,
-0.062347412109375,
-0.047271728515625,
-0.0120697021484375,
-0.017974853515625,
-0.046630859375,
0.0272064208984375,
-0.01253509521484375,
-0.005176544189453125,
0.0114288330078125,
0.0004200935363769531,
-0.00371551513671875,
-0.005817413330078125,
0.034210205078125,
0.01346588134765625,
-0.01007843017578125,
-0.01300048828125,
0.0089263916015625,
0.01245880126953125,
-0.0033168792724609375,
-0.01409149169921875,
0.037872314453125,
-0.0185394287109375,
-0.00787353515625,
-0.0309600830078125,
0.0217437744140625,
0.03717041015625,
-0.00015485286712646484,
0.07196044921875,
0.0596923828125,
-0.0261688232421875,
0.020599365234375,
-0.0369873046875,
-0.01172637939453125,
-0.037078857421875,
0.01041412353515625,
-0.0193023681640625,
-0.0657958984375,
0.04913330078125,
0.0347900390625,
0.0189361572265625,
0.052520751953125,
0.03765869140625,
-0.0002868175506591797,
0.07586669921875,
0.036407470703125,
-0.01995849609375,
0.0338134765625,
-0.033843994140625,
-0.01074981689453125,
-0.062347412109375,
-0.0199432373046875,
-0.025421142578125,
-0.015106201171875,
-0.056060791015625,
-0.03900146484375,
0.0128021240234375,
0.03143310546875,
-0.04815673828125,
0.04119873046875,
-0.04718017578125,
0.01629638671875,
0.037353515625,
0.01535797119140625,
0.0152130126953125,
-0.0035343170166015625,
-0.0164031982421875,
0.0034160614013671875,
-0.04412841796875,
-0.0256195068359375,
0.07977294921875,
0.0303802490234375,
0.04443359375,
0.0281524658203125,
0.03515625,
0.006465911865234375,
0.032623291015625,
-0.032958984375,
0.047027587890625,
0.00870513916015625,
-0.058624267578125,
-0.0311279296875,
-0.047637939453125,
-0.059112548828125,
0.018463134765625,
-0.010040283203125,
-0.06982421875,
0.018157958984375,
0.00027751922607421875,
-0.0165252685546875,
0.020965576171875,
-0.040252685546875,
0.079833984375,
-0.00888824462890625,
-0.0218048095703125,
-0.01113128662109375,
-0.05853271484375,
0.0380859375,
0.00861358642578125,
0.00811767578125,
-0.025909423828125,
-0.00930023193359375,
0.05950927734375,
-0.0645751953125,
0.06561279296875,
-0.0213470458984375,
0.005962371826171875,
0.040924072265625,
-0.0025577545166015625,
0.03515625,
0.0216064453125,
-0.0002925395965576172,
0.0279388427734375,
0.037445068359375,
-0.036834716796875,
-0.02325439453125,
0.042510986328125,
-0.08074951171875,
-0.033172607421875,
-0.03173828125,
-0.028594970703125,
-0.00045228004455566406,
0.002742767333984375,
0.0281829833984375,
0.0289459228515625,
-0.0159759521484375,
0.0182037353515625,
0.038482666015625,
-0.0389404296875,
0.02178955078125,
0.0294342041015625,
-0.03271484375,
-0.0467529296875,
0.04937744140625,
-0.004241943359375,
0.02490234375,
0.0204315185546875,
0.0007963180541992188,
-0.0298614501953125,
-0.0263519287109375,
-0.040863037109375,
0.034088134765625,
-0.037872314453125,
-0.031585693359375,
-0.054595947265625,
-0.0302581787109375,
-0.036285400390625,
0.0223846435546875,
-0.0289459228515625,
-0.0496826171875,
-0.041778564453125,
0.0033359527587890625,
0.07110595703125,
0.03155517578125,
-0.0274810791015625,
0.0251617431640625,
-0.0645751953125,
0.0104827880859375,
0.03204345703125,
0.01885986328125,
-0.006256103515625,
-0.053863525390625,
-0.0013780593872070312,
0.03204345703125,
-0.036529541015625,
-0.073486328125,
0.043853759765625,
0.026947021484375,
0.0396728515625,
0.0247344970703125,
0.01080322265625,
0.062164306640625,
-0.0139617919921875,
0.076171875,
0.0213623046875,
-0.0679931640625,
0.040771484375,
-0.039093017578125,
0.007785797119140625,
0.0308837890625,
0.04534912109375,
-0.03289794921875,
-0.0274810791015625,
-0.056671142578125,
-0.051727294921875,
0.0447998046875,
0.03570556640625,
0.0152130126953125,
0.0036792755126953125,
0.0433349609375,
-0.00818634033203125,
0.00787353515625,
-0.0650634765625,
-0.04144287109375,
-0.03228759765625,
-0.00327301025390625,
0.0101470947265625,
-0.0152130126953125,
-0.017913818359375,
-0.05145263671875,
0.06329345703125,
-0.012603759765625,
0.05364990234375,
0.0147552490234375,
0.01239013671875,
-0.01378631591796875,
0.004283905029296875,
0.03729248046875,
0.0419921875,
-0.02166748046875,
-0.0187530517578125,
0.0215301513671875,
-0.04736328125,
0.001560211181640625,
0.01806640625,
0.0086822509765625,
-0.01035308837890625,
0.01177215576171875,
0.0626220703125,
0.005245208740234375,
-0.029754638671875,
0.050384521484375,
-0.026702880859375,
-0.031707763671875,
-0.01483154296875,
0.00937652587890625,
0.027740478515625,
0.03515625,
0.0222930908203125,
-0.0243377685546875,
0.0139007568359375,
-0.04388427734375,
0.01541900634765625,
0.047119140625,
-0.01531219482421875,
-0.03692626953125,
0.055267333984375,
-0.006877899169921875,
0.004009246826171875,
0.053558349609375,
-0.0227508544921875,
-0.029754638671875,
0.049835205078125,
0.036102294921875,
0.043670654296875,
-0.01837158203125,
0.0250396728515625,
0.0401611328125,
0.01556396484375,
-0.00185394287109375,
0.036834716796875,
-0.003719329833984375,
-0.048980712890625,
-0.035491943359375,
-0.047607421875,
-0.0310211181640625,
0.0220489501953125,
-0.056671142578125,
0.01177978515625,
-0.030029296875,
-0.0357666015625,
-0.005771636962890625,
0.020721435546875,
-0.037200927734375,
0.00891876220703125,
0.0031948089599609375,
0.0748291015625,
-0.052703857421875,
0.06817626953125,
0.046112060546875,
-0.0295562744140625,
-0.06915283203125,
-0.0274200439453125,
0.01025390625,
-0.04547119140625,
0.0211334228515625,
-0.002620697021484375,
0.024505615234375,
-0.006122589111328125,
-0.06329345703125,
-0.06854248046875,
0.1036376953125,
0.0208892822265625,
-0.04180908203125,
-0.004283905029296875,
-0.00307464599609375,
0.0295562744140625,
-0.013916015625,
0.046661376953125,
0.035552978515625,
0.0229034423828125,
0.0227508544921875,
-0.09185791015625,
0.0280303955078125,
-0.033111572265625,
-0.00806427001953125,
0.0120086669921875,
-0.08074951171875,
0.0718994140625,
-0.00920867919921875,
-0.0141448974609375,
0.0093994140625,
0.048858642578125,
0.029327392578125,
-0.0016918182373046875,
0.033233642578125,
0.052825927734375,
0.04913330078125,
-0.01995849609375,
0.0743408203125,
-0.01149749755859375,
0.045013427734375,
0.05950927734375,
0.0025005340576171875,
0.06201171875,
0.0165252685546875,
-0.051483154296875,
0.043853759765625,
0.07501220703125,
0.00678253173828125,
0.032958984375,
-0.005947113037109375,
-0.01806640625,
-0.00872802734375,
0.00875091552734375,
-0.05120849609375,
0.010498046875,
0.0391845703125,
-0.013824462890625,
0.0099029541015625,
-0.01215362548828125,
0.00518798828125,
-0.054962158203125,
-0.0011453628540039062,
0.0516357421875,
0.0287017822265625,
-0.01065826416015625,
0.06854248046875,
-0.012542724609375,
0.0482177734375,
-0.03521728515625,
-0.00655364990234375,
-0.0282440185546875,
-0.006328582763671875,
-0.020355224609375,
-0.05963134765625,
0.0254364013671875,
-0.0204315185546875,
0.00734710693359375,
0.00824737548828125,
0.050140380859375,
-0.014862060546875,
-0.01690673828125,
0.0296783447265625,
0.034576416015625,
0.0307159423828125,
-0.00376129150390625,
-0.08642578125,
0.0256500244140625,
0.003482818603515625,
-0.0521240234375,
0.035888671875,
0.0311279296875,
0.009185791015625,
0.0548095703125,
0.04547119140625,
-0.005641937255859375,
-0.00920867919921875,
-0.0162353515625,
0.0823974609375,
-0.052642822265625,
-0.0184173583984375,
-0.06573486328125,
0.041015625,
-0.0189666748046875,
-0.024200439453125,
0.05560302734375,
0.039459228515625,
0.047821044921875,
0.015655517578125,
0.04815673828125,
-0.036285400390625,
0.0213623046875,
-0.0163116455078125,
0.048095703125,
-0.054443359375,
0.0089569091796875,
-0.02764892578125,
-0.05438232421875,
-0.01076507568359375,
0.0579833984375,
-0.01080322265625,
0.0119781494140625,
0.0243988037109375,
0.06304931640625,
-0.0042266845703125,
0.01316070556640625,
0.01416778564453125,
0.0260467529296875,
0.0150146484375,
0.06591796875,
0.062469482421875,
-0.06500244140625,
0.04669189453125,
-0.0311737060546875,
-0.020050048828125,
-0.0189361572265625,
-0.062744140625,
-0.07135009765625,
-0.03668212890625,
-0.049896240234375,
-0.04833984375,
-0.0015811920166015625,
0.06573486328125,
0.0704345703125,
-0.048797607421875,
-0.01959228515625,
0.01004791259765625,
0.0022525787353515625,
-0.01910400390625,
-0.02203369140625,
0.0196533203125,
0.029815673828125,
-0.0335693359375,
0.006305694580078125,
0.01123809814453125,
0.032989501953125,
-0.0126800537109375,
-0.0288543701171875,
-0.0194854736328125,
0.00914764404296875,
0.046539306640625,
0.0479736328125,
-0.044921875,
-0.0193939208984375,
-0.0034275054931640625,
-0.0185546875,
0.016265869140625,
0.01137542724609375,
-0.0472412109375,
-0.004581451416015625,
0.03668212890625,
0.002613067626953125,
0.0694580078125,
0.0047607421875,
0.02197265625,
-0.04315185546875,
0.017669677734375,
0.0015125274658203125,
0.018585205078125,
0.0176544189453125,
-0.0306243896484375,
0.04974365234375,
0.0239410400390625,
-0.0504150390625,
-0.06585693359375,
-0.014556884765625,
-0.09124755859375,
-0.01093292236328125,
0.0877685546875,
-0.016754150390625,
-0.022613525390625,
-0.00193023681640625,
-0.0290374755859375,
0.02288818359375,
-0.035919189453125,
0.01461029052734375,
0.038330078125,
-0.021514892578125,
-0.03741455078125,
-0.05718994140625,
0.048675537109375,
0.008392333984375,
-0.061248779296875,
0.0017137527465820312,
0.038238525390625,
0.03961181640625,
0.0005445480346679688,
0.0853271484375,
-0.0217742919921875,
0.0223236083984375,
0.003787994384765625,
0.002300262451171875,
0.007213592529296875,
0.009307861328125,
-0.01513671875,
-0.010589599609375,
-0.0083465576171875,
-0.002094268798828125
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.