Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
gemini25flash-stackexchange-overflow-32ep-512k-v3-traces – AI Model by DCAgent2 | AlphaNeural AI
You can deploy this model and start earning money today!
DCAgent2
/
gemini25flash-stackexchange-overflow-32ep-512k-v3-traces
like
0
transformers
safetensors
qwen3
text-generation
llama-factory
full
generated_from_trainer
conversational
Qwen/Qwen3-8B
finetune
apache-2.0
text-generation-inference
endpoints_compatible
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
gemini25flash-stackexchange-overflow-32ep-512k-v3-traces
This model is a fine-tuned version of
Qwen/Qwen3-8B
on the DCAgent2/gemini25flash-stackexchange-overflow-32ep-512k-v3-traces 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: 4e-05
train_batch_size: 1
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 16
total_train_batch_size: 16
total_eval_batch_size: 128
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_ratio: 0.1
num_epochs: 7.0
Training results
Framework versions
Transformers 4.56.0
Pytorch 2.7.0+cu128
Datasets 4.4.1
Tokenizers 0.22.1
