unconst/Affine-5czsc2fc98-r478-offline-dpo-hialpha-extrasteps-merged
Affine-5czsc2fc98-r478-offline-dpo-hialpha-extrasteps-merged is a 35.1 billion parameter language model, merged from kevin954/Affine-5dfqbbh8ev-sft. This model has a context length of 32768 tokens. It is a private checkpoint intended for TTL insurance, not yet a public submission.
Loading preview...
Model Overview
This model, named Affine-5czsc2fc98-r478-offline-dpo-hialpha-extrasteps-merged, is a large language model with 35.1 billion parameters and a 32768 token context length. It was created by unconst and is a merged checkpoint, specifically derived from kevin954/Affine-5dfqbbh8ev-sft.
Key Characteristics
- Parameter Count: 35.1 billion, indicating a substantial capacity for complex language understanding and generation.
- Context Length: 32768 tokens, allowing it to process and generate longer sequences of text, which is beneficial for tasks requiring extensive context.
- Origin: It is a LoRA-merged checkpoint, suggesting it has undergone fine-tuning or adaptation from a base model.
Current Status
This particular version is described as a "private TTL insurance" checkpoint, meaning it is not yet intended for public submission or general use. It is awaiting a "Stage-5 gate clears" before it can be considered a full submission. Developers should note its developmental status and understand it is not a finalized public release.