counseLLM API & Inference Endpoint

Model Details

Developed by: Gowtham Arulmozhii
Model type: Causal Language Model (text generation)
Language: English
License: Apache 2.0
Base model: meta-llama/Llama-3.1-8B-Instruct
Repository: GitHub

Training

Two-Stage Alignment Pipeline

Stage 1 — Supervised Fine-Tuning (SFT)

Table with columns: Parameter, Value
Parameter	Value
Method	QLoRA (4-bit NF4 + double quantization)
LoRA Rank / Alpha	64 / 128
Learning Rate	2e-4 (cosine scheduler)
Epochs	2
Effective Batch Size	16
Training Data	36K multi-source counseling examples
GPU	NVIDIA H100 80GB
Training Time	~3 hours

Stage 2 — Direct Preference Optimization (DPO)

Table with columns: Parameter, Value
Parameter	Value
Method	QLoRA on SFT-merged base
LoRA Rank / Alpha	16 / 32
Beta (KL penalty)	0.5
Learning Rate	1e-5 (cosine scheduler)
Epochs	1
Effective Batch Size	8
Training Data	~2K preference-filtered pairs
GPU	NVIDIA H100 80GB
Training Time	~30 minutes

Training Data

SFT (36K examples from 5 sources)

Table with columns: Source, Examples, Type
Source	Examples	Type
MentalChat16K	~16K	Synthetic + clinical
empathetic_dialogues	~10K	Real human multi-turn
Psych8k	~8K	Real therapist transcripts
counsel-chat

DPO (~2K preference pairs)

Table with columns: Source, Pairs, Selection
Source	Pairs	Selection
PsychoCounsel-Preference	~2K	Rating-gap filtered across 7 dimensions

Evaluation

Automated Metrics

Table with columns: Metric, Base, SFT, DPO
Metric	Base	SFT	DPO
Perplexity	4.18	3.64	3.13
BERTScore F1	0.8598	0.8527	0.8492
ROUGE-L F1	0.1065	0.0772	0.0790
Distinct-1	0.273	0.331	0.262

LLM-as-Judge (GPT-4o, 1-5 scale)

Table with columns: Dimension, Base, SFT, DPO
Dimension	Base	SFT	DPO
Empathy	4.40	3.48	4.88
Safety	4.28	3.84	4.60
Relevance	4.68	3.72	4.88
Helpfulness	4.04	3.04

Evaluated on 25 curated prompts across 18 mental health categories (anxiety, depression, grief, crisis, relationships, trauma, etc.).

Available Checkpoints

This repo contains three artifacts:

Table with columns: Path, Format, Size, Description
Path	Format	Size	Description
`/` (root)	Full merged model	~16 GB	Ready-to-use Llama 3.1 8B + SFT + DPO merged
`sft/`	LoRA adapter	~640 MB	Stage-1 SFT adapter (r=64, α=128) — load on top of base Llama 3.1 8B
`dpo/`	LoRA adapter	~160 MB	Stage-2 DPO adapter (r=16, α=32) — load on top of SFT-merged base

Load an adapter with PEFT:

python
from peft import PeftModel
from transformers import AutoModelForCausalLM

base = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct", torch_dtype="bfloat16", device_map="auto")
model = PeftModel.from_pretrained(base, "Wothmag07/counseLLM", subfolder="dpo")

How to Use

python
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "Wothmag07/counseLLM"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

messages = [
    {"role": "system", "content": "You are a mental health counselor providing supportive, empathetic guidance. Respond by first acknowledging the person's feelings, then explore their situation with open-ended questions. Use techniques like reflective listening, validation, and gentle reframing. Keep responses warm, conversational, and non-judgmental."},
    {"role": "user", "content": "I've been feeling really anxious about work lately and I can't sleep."},
]

input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
outputs = model.generate(
    input_ids,
    max_new_tokens=512,
    temperature=0.7,
    top_p=0.9,
    repetition_penalty=1.1,
)
response = tokenizer.decode(outputs[0][input_ids.shape[1]:], skip_special_tokens=True)
print(response)

Uses

Intended Use

Research and educational purposes in AI-assisted mental health support
Studying alignment techniques (SFT + DPO) applied to sensitive domains
Demonstrating empathy-aligned language model fine-tuning

Out-of-Scope Use

Clinical deployment — this model is not validated for clinical use
Crisis intervention — should not be relied upon for suicide prevention or emergency situations
Replacement for therapy — not a substitute for licensed mental health professionals

Bias, Risks, and Limitations

The model may reflect biases present in training data (both real and synthetic sources)
Responses may sometimes be generic or miss nuances of specific cultural contexts
The model may generate plausible-sounding but clinically inaccurate advice
Training data is predominantly English and may not generalize to other languages
Should not be deployed in production clinical settings without extensive safety review

Environmental Impact

Hardware: NVIDIA H100 80GB
Training Time: ~3.5 hours total (SFT: 3h, DPO: 30min)
Cloud Provider: Modal

Tech Stack

Table with columns: Component, Technology
Component	Technology
Base Model	Meta Llama 3.1 8B Instruct
Training	HuggingFace TRL (SFTTrainer, DPOTrainer)
Quantization	QLoRA via bitsandbytes (4-bit NF4)
Adapters	PEFT (LoRA)
Infrastructure	Modal (H100 GPUs)
Experiment Tracking	Weights & Biases
Evaluation	BERTScore, ROUGE-L, GPT-4o Judge

Citation

bibtex
@misc{counseLLM2026,
  author = {Gowtham Arulmozhii},
  title = {CounseLLM: Empathy-Aligned Conversational Support LLM},
  year = {2026},
  publisher = {HuggingFace},
  url = {https://huggingface.co/Wothmag07/counseLLM}
}

Model Card Contact

GitHub: @wothmag07

Parameter

Value

Method

QLoRA (4-bit NF4 + double quantization)

LoRA Rank / Alpha

64 / 128

Learning Rate

2e-4 (cosine scheduler)

Epochs

Effective Batch Size

Training Data

36K multi-source counseling examples

GPU

NVIDIA H100 80GB

Training Time

~3 hours

Parameter

Value

Method

QLoRA on SFT-merged base

LoRA Rank / Alpha

16 / 32

Beta (KL penalty)

0.5

Learning Rate

1e-5 (cosine scheduler)

Epochs

Effective Batch Size

Training Data

~2K preference-filtered pairs

GPU

NVIDIA H100 80GB

Training Time

~30 minutes

Source

Examples

Type

MentalChat16K

~16K

Synthetic + clinical

empathetic_dialogues

~10K

Real human multi-turn

Psych8k

~8K

Real therapist transcripts

counsel-chat

Source

Pairs

Selection

PsychoCounsel-Preference

~2K

Rating-gap filtered across 7 dimensions

Metric

Base

SFT

DPO

Perplexity

4.18

3.64

3.13

BERTScore F1

0.8598

0.8527

0.8492

ROUGE-L F1

0.1065

0.0772

0.0790

Distinct-1

0.273

0.331

0.262

Dimension

Base

SFT

DPO

Empathy

4.40

3.48

4.88

Safety

4.28

3.84

4.60

Relevance

4.68

3.72

4.88

Helpfulness

4.04

3.04

Path

Format

Size

Description

/ (root)

Full merged model

~16 GB

Ready-to-use Llama 3.1 8B + SFT + DPO merged

sft/

LoRA adapter

~640 MB

Stage-1 SFT adapter (r=64, α=128) — load on top of base Llama 3.1 8B

dpo/

LoRA adapter

~160 MB

Stage-2 DPO adapter (r=16, α=32) — load on top of SFT-merged base

python

from peft import PeftModel
from transformers import AutoModelForCausalLM

base = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-3.1-8B-Instruct", torch_dtype="bfloat16", device_map="auto")
model = PeftModel.from_pretrained(base, "Wothmag07/counseLLM", subfolder="dpo")

python

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "Wothmag07/counseLLM"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

messages = [
    {"role": "system", "content": "You are a mental health counselor providing supportive, empathetic guidance. Respond by first acknowledging the person's feelings, then explore their situation with open-ended questions. Use techniques like reflective listening, validation, and gentle reframing. Keep responses warm, conversational, and non-judgmental."},
    {"role": "user", "content": "I've been feeling really anxious about work lately and I can't sleep."},
]

input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)
outputs = model.generate(
    input_ids,
    max_new_tokens=512,
    temperature=0.7,
    top_p=0.9,
    repetition_penalty=1.1,
)
response = tokenizer.decode(outputs[0][input_ids.shape[1]:], skip_special_tokens=True)
print(response)

Component

Technology

Base Model

Meta Llama 3.1 8B Instruct

Training

HuggingFace TRL (SFTTrainer, DPOTrainer)

Quantization

QLoRA via bitsandbytes (4-bit NF4)

Adapters

PEFT (LoRA)

Infrastructure

Modal (H100 GPUs)

Experiment Tracking

Weights & Biases

Evaluation

BERTScore, ROUGE-L, GPT-4o Judge

bibtex

@misc{counseLLM2026,
  author = {Gowtham Arulmozhii},
  title = {CounseLLM: Empathy-Aligned Conversational Support LLM},
  year = {2026},
  publisher = {HuggingFace},
  url = {https://huggingface.co/Wothmag07/counseLLM}
}

counseLLM

README

Model Details

Training

Two-Stage Alignment Pipeline

Training Data

Evaluation

Automated Metrics

LLM-as-Judge (GPT-4o, 1-5 scale)

Available Checkpoints

How to Use

Uses

Intended Use

Out-of-Scope Use

Bias, Risks, and Limitations

Environmental Impact

Tech Stack

Citation

Model Card Contact

Explore FriendliAI today

README

Model Details

Training

Two-Stage Alignment Pipeline

Training Data

Evaluation

Automated Metrics

LLM-as-Judge (GPT-4o, 1-5 scale)

Available Checkpoints

How to Use

Uses

Intended Use

Out-of-Scope Use

Bias, Risks, and Limitations

Environmental Impact

Tech Stack

Citation

Model Card Contact