---
language:
- en
license: apache-2.0
library_name: transformers
tags:
- medical
- trl
- trainer
datasets:
- ShieldX/manovyadh-3.5k
metrics:
- accuracy
thumbnail: https://huggingface.co/ShieldX/manovyadh-1.1B-v1-chat/blob/main/manovyadh.png
pipeline_tag: text-generation
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
widget:
- text: '###SYSTEM: You are an AI assistant that helps people cope with stress and
improve their mental health. User will tell you about their feelings and challenges.
Your task is to listen empathetically and offer helpful suggestions. While responding,
think about the user’s needs and goals and show compassion and support
###USER: I don''t know how to tell someone how I feel about them. How can I get
better at expressing how I feel??
###ASSISTANT:
'
model-index:
- name: manovyadh-1.1B-v1-chat
results:
- task:
type: text-generation
dataset:
name: ai2_arc
type: arc
metrics:
- type: pass@1
value: 35.92
name: pass@1
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
name: Open LLM Leaderboard
- task:
type: text-generation
dataset:
name: hellaswag
type: hellaswag
metrics:
- type: pass@1
value: 60.03
name: pass@1
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
name: Open LLM Leaderboard
- task:
type: text-generation
dataset:
name: truthful_qa
type: truthful_qa
metrics:
- type: pass@1
value: 39.17
name: pass@1
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
name: Open LLM Leaderboard
- task:
type: text-generation
dataset:
name: winogrande
type: winogrande
metrics:
- type: pass@1
value: 61.09
name: pass@1
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: AI2 Reasoning Challenge (25-Shot)
type: ai2_arc
config: ARC-Challenge
split: test
args:
num_few_shot: 25
metrics:
- type: acc_norm
value: 35.92
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ShieldX/manovyadh-1.1B-v1-chat
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: HellaSwag (10-Shot)
type: hellaswag
split: validation
args:
num_few_shot: 10
metrics:
- type: acc_norm
value: 60.03
name: normalized accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ShieldX/manovyadh-1.1B-v1-chat
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU (5-Shot)
type: cais/mmlu
config: all
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 25.82
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ShieldX/manovyadh-1.1B-v1-chat
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: TruthfulQA (0-shot)
type: truthful_qa
config: multiple_choice
split: validation
args:
num_few_shot: 0
metrics:
- type: mc2
value: 39.17
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ShieldX/manovyadh-1.1B-v1-chat
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: Winogrande (5-shot)
type: winogrande
config: winogrande_xl
split: validation
args:
num_few_shot: 5
metrics:
- type: acc
value: 61.09
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ShieldX/manovyadh-1.1B-v1-chat
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GSM8k (5-shot)
type: gsm8k
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 1.74
name: accuracy
source:
url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ShieldX/manovyadh-1.1B-v1-chat
name: Open LLM Leaderboard
---
# Uploaded model
- **Developed by:** ShieldX
- **License:** apache-2.0
- **Finetuned from model :** TinyLlama/TinyLlama-1.1B-Chat-v1.0
# ShieldX/manovyadh-1.1B-v1
Introducing ManoVyadh, A finetuned version of TinyLlama 1.1B Chat on Mental Health Counselling Dataset.
# Model Details
## Model Description
ManoVyadh is a LLM for mental health counselling.
# Uses
## Direct Use
- base model for further finetuning
- for fun
## Downstream Use
- can be deployed with api
- used to create webapp or app to show demo
## Out-of-Scope Use
- cannot be used for production purpose
- not to be applied in real life health purpose
- cannot be used to generate text for research or academic purposes
# Usage
```
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM, AutoConfig
tokenizer = AutoTokenizer.from_pretrained("ShieldX/manovyadh-1.1B-v1-chat")
model = AutoModelForCausalLM.from_pretrained("ShieldX/manovyadh-1.1B-v1-chat").to("cuda")
config = AutoConfig.from_pretrained("ShieldX/manovyadh-1.1B-v1-chat")
def format_prompt(q):
return f"""###SYSTEM: You are an AI assistant that helps people cope with stress and improve their mental health. User will tell you about their feelings and challenges. Your task is to listen empathetically and offer helpful suggestions. While responding, think about the user’s needs and goals and show compassion and support
###USER: {q}
###ASSISTANT:"""
prompt = format_prompt("I've never been able to talk with my parents. My parents are in their sixties while I am a teenager. I love both of them but not their personalities. I feel that they do not take me seriously whenever I talk about a serious event in my life. If my dad doesn’t believe me, then my mom goes along with my dad and acts like she doesn’t believe me either. I’m a pansexual, but I can’t trust my own parents. I've fought depression and won; however, stress and anxiety are killing me. I feel that my friends don't listen to me. I know they have their own problems, which I do my best to help with. But they don't always try to help me with mine, when I really need them. I feel as if my childhood has been taken from me. I feel as if I have no one whom I can trust.")
import torch
from transformers import GenerationConfig, TextStreamer
from time import perf_counter
# Check for GPU availability
if torch.cuda.is_available():
device = "cuda"
else:
device = "cpu"
# Move model and inputs to the GPU (if available)
model.to(device)
inputs = tokenizer(prompt, return_tensors="pt").to(device)
streamer = TextStreamer(tokenizer)
generation_config = GenerationConfig(
penalty_alpha=0.6,
do_sample=True,
top_k=5,
temperature=0.5,
repetition_penalty=1.2,
max_new_tokens=256,
streamer=streamer,
pad_token_id=tokenizer.eos_token_id
)
start_time = perf_counter()
outputs = model.generate(**inputs, generation_config=generation_config)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
output_time = perf_counter() - start_time
print(f"Time taken for inference: {round(output_time, 2)} seconds")
```
# 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.
# Training Details
# Model Examination
We will be further finetuning this model on large dataset to see how it performs
# 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:** 1 X Tesla T4
- **Hours used:** 0.48
- **Cloud Provider:** Google Colab
- **Compute Region:** India
# Technical Specifications
## Model Architecture and Objective
Finetuned on Tiny-Llama 1.1B Chat model
### Hardware
1 X Tesla T4
# training
This model is a fine-tuned version of [TinyLlama/TinyLlama-1.1B-Chat-v1.0](https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v1.0) on [ShieldX/manovyadh-3.5k](https://huggingface.co/datasets/ShieldX/manovyadh-3.5k) dataset.
It achieves the following results on the evaluation set:
- Loss: 1.8587
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2.5e-05
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- training_steps: 400
- mixed_precision_training: Native AMP
-
### Training results
| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 2.5894 | 0.01 | 5 | 2.5428 |
| 2.5283 | 0.02 | 10 | 2.5240 |
| 2.5013 | 0.03 | 15 | 2.5033 |
| 2.378 | 0.05 | 20 | 2.4770 |
| 2.3735 | 0.06 | 25 | 2.4544 |
| 2.3894 | 0.07 | 30 | 2.4335 |
| 2.403 | 0.08 | 35 | 2.4098 |
| 2.3719 | 0.09 | 40 | 2.3846 |
| 2.3691 | 0.1 | 45 | 2.3649 |
| 2.3088 | 0.12 | 50 | 2.3405 |
| 2.3384 | 0.13 | 55 | 2.3182 |
| 2.2577 | 0.14 | 60 | 2.2926 |
| 2.245 | 0.15 | 65 | 2.2702 |
| 2.1389 | 0.16 | 70 | 2.2457 |
| 2.1482 | 0.17 | 75 | 2.2176 |
| 2.1567 | 0.18 | 80 | 2.1887 |
| 2.1533 | 0.2 | 85 | 2.1616 |
| 2.0629 | 0.21 | 90 | 2.1318 |
| 2.1068 | 0.22 | 95 | 2.0995 |
| 2.0196 | 0.23 | 100 | 2.0740 |
| 2.062 | 0.24 | 105 | 2.0461 |
| 1.9436 | 0.25 | 110 | 2.0203 |
| 1.9348 | 0.26 | 115 | 1.9975 |
| 1.8803 | 0.28 | 120 | 1.9747 |
| 1.9108 | 0.29 | 125 | 1.9607 |
| 1.7826 | 0.3 | 130 | 1.9506 |
| 1.906 | 0.31 | 135 | 1.9374 |
| 1.8745 | 0.32 | 140 | 1.9300 |
| 1.8634 | 0.33 | 145 | 1.9232 |
| 1.8561 | 0.35 | 150 | 1.9183 |
| 1.8371 | 0.36 | 155 | 1.9147 |
| 1.8006 | 0.37 | 160 | 1.9106 |
| 1.8941 | 0.38 | 165 | 1.9069 |
| 1.8456 | 0.39 | 170 | 1.9048 |
| 1.8525 | 0.4 | 175 | 1.9014 |
| 1.8475 | 0.41 | 180 | 1.8998 |
| 1.8255 | 0.43 | 185 | 1.8962 |
| 1.9358 | 0.44 | 190 | 1.8948 |
| 1.758 | 0.45 | 195 | 1.8935 |
| 1.7859 | 0.46 | 200 | 1.8910 |
| 1.8412 | 0.47 | 205 | 1.8893 |
| 1.835 | 0.48 | 210 | 1.8875 |
| 1.8739 | 0.49 | 215 | 1.8860 |
| 1.9397 | 0.51 | 220 | 1.8843 |
| 1.8187 | 0.52 | 225 | 1.8816 |
| 1.8174 | 0.53 | 230 | 1.8807 |
| 1.8 | 0.54 | 235 | 1.8794 |
| 1.7736 | 0.55 | 240 | 1.8772 |
| 1.7429 | 0.56 | 245 | 1.8778 |
| 1.8024 | 0.58 | 250 | 1.8742 |
| 1.8431 | 0.59 | 255 | 1.8731 |
| 1.7692 | 0.6 | 260 | 1.8706 |
| 1.8084 | 0.61 | 265 | 1.8698 |
| 1.7602 | 0.62 | 270 | 1.8705 |
| 1.7751 | 0.63 | 275 | 1.8681 |
| 1.7403 | 0.64 | 280 | 1.8672 |
| 1.8078 | 0.66 | 285 | 1.8648 |
| 1.8464 | 0.67 | 290 | 1.8648 |
| 1.7853 | 0.68 | 295 | 1.8651 |
| 1.8546 | 0.69 | 300 | 1.8643 |
| 1.8319 | 0.7 | 305 | 1.8633 |
| 1.7908 | 0.71 | 310 | 1.8614 |
| 1.738 | 0.72 | 315 | 1.8625 |
| 1.8868 | 0.74 | 320 | 1.8630 |
| 1.7744 | 0.75 | 325 | 1.8621 |
| 1.8292 | 0.76 | 330 | 1.8609 |
| 1.7905 | 0.77 | 335 | 1.8623 |
| 1.7652 | 0.78 | 340 | 1.8610 |
| 1.8371 | 0.79 | 345 | 1.8611 |
| 1.7024 | 0.81 | 350 | 1.8593 |
| 1.7328 | 0.82 | 355 | 1.8593 |
| 1.7376 | 0.83 | 360 | 1.8606 |
| 1.747 | 0.84 | 365 | 1.8601 |
| 1.7777 | 0.85 | 370 | 1.8602 |
| 1.8701 | 0.86 | 375 | 1.8598 |
| 1.7165 | 0.87 | 380 | 1.8579 |
| 1.779 | 0.89 | 385 | 1.8588 |
| 1.8536 | 0.9 | 390 | 1.8583 |
| 1.7263 | 0.91 | 395 | 1.8582 |
| 1.7983 | 0.92 | 400 | 1.8587 |
### Framework versions
- PEFT 0.7.1
- Transformers 4.37.1
- Pytorch 2.1.0+cu121
- Datasets 2.16.1
- Tokenizers 0.15.1
# Citation
**BibTeX:**
```
@misc{ShieldX/manovyadh-1.1B-v1-chat,
url={[https://huggingface.co/ShieldX/manovyadh-1.1B-v1-chat](https://huggingface.co/ShieldX/manovyadh-1.1B-v1-chat)},
title={ManoVyadh},
author={Rohan Shaw},
year={2024}, month={Jan}
}
```
# Model Card Authors
ShieldX a.k.a Rohan Shaw
# Model Card Contact
email : rohanshaw.dev@gmail.com
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_ShieldX__manovyadh-1.1B-v1-chat)
| Metric |Value|
|---------------------------------|----:|
|Avg. |37.30|
|AI2 Reasoning Challenge (25-Shot)|35.92|
|HellaSwag (10-Shot) |60.03|
|MMLU (5-Shot) |25.82|
|TruthfulQA (0-shot) |39.17|
|Winogrande (5-shot) |61.09|
|GSM8k (5-shot) | 1.74|