This is a very small randomly initialized MistralForCausalLM model created by the TRL (Transformers Reinforcement Learning) team for internal testing. It is used to validate training scripts, inference pipelines, and integration tests without the overhead of loading full-scale models. Not intended for actual language generation.
In the Foundation models & chat space, Tiny MistralForCausalLM takes a focused approach. It focuses on providing minimal test models for validating Mistral architecture integration and training pipelines. It is built as an open-source project for developers. Tiny MistralForCausalLM is open source under the Apache-2.0 license. It runs on the web, the command line, and API.
Behind Tiny MistralForCausalLM is trl-internal-testing, and the product first shipped in 2020. Development happens publicly on GitHub with 18.9k stars and 521 commits in the last 90 days. Key capabilities include causal language model, test model, and random weights.
Latest indexed changes and source events
trl-internal-testing/tiny-MistralForCausalLM-0.2 verified by the PulseGate indexer
