g-assismoraes/DeltaP2S-Llama2-13B-P2S-CodeLlama7B-SameFormula-S13-QV

TEXT GENERATIONPricing:Input $1.5 / Output $2.1Concurrent Unit Cost:1Model Size:13BQuant:FP8Context Size:4kPublished:Aug 29, 2026Architecture:Transformer Featherless Exclusive Cold

The g-assismoraes/DeltaP2S-Llama2-13B-P2S-CodeLlama7B-SameFormula-S13-QV is a 13 billion parameter merged checkpoint model, derived from a Delta-P2S experiment. This model integrates components from Llama2-13B and CodeLlama7B, specifically trained using the 'SameFormula-S13-QV' configuration. It is designed as a specialized merge for specific research or application contexts, focusing on the outcomes of the Delta-P2S training methodology.

Loading preview...

Model Overview

The g-assismoraes/DeltaP2S-Llama2-13B-P2S-CodeLlama7B-SameFormula-S13-QV is a 13 billion parameter language model representing a merged checkpoint from a specific Delta-P2S experiment. This model is a result of combining elements from a Llama2-13B base with a CodeLlama7B component, indicating a potential focus on tasks that benefit from both general language understanding and code-specific capabilities.

Key Characteristics

  • Architecture: Merged checkpoint from a Delta-P2S experiment.
  • Base Models: Integrates components from Llama2-13B and CodeLlama7B.
  • Parameter Count: 13 billion parameters.
  • Training Configuration: Utilizes the 'SameFormula-S13-QV' training setup, suggesting a specialized fine-tuning or merging strategy.

Intended Use Cases

This model is primarily suited for researchers and developers interested in:

  • Exploring the outcomes and performance of the Delta-P2S merging methodology.
  • Applications requiring a blend of Llama2's general language understanding and CodeLlama's code-centric strengths.
  • Specific tasks aligned with the 'SameFormula-S13-QV' training objectives, which would need further context from the original experiment documentation.

Finetunes
1 models