fe6e9c4862
Model: namkoong-lab/LatentGym_Qwen3-8B_10episodes_MultiLatent_number_guessing Source: Original Platform
46 lines
1.1 KiB
Markdown
46 lines
1.1 KiB
Markdown
---
|
|
base_model: Qwen/Qwen3-8B
|
|
license: apache-2.0
|
|
library_name: transformers
|
|
pipeline_tag: text-generation
|
|
language: [en]
|
|
tags:
|
|
- qwen3
|
|
- meta-rl
|
|
- in-context-learning
|
|
- grpo
|
|
---
|
|
|
|
# LatentGym_Qwen3-8B_10episodes_MultiLatent_number_guessing
|
|
|
|
GRPO-fine-tuned Qwen3-8B for in-context meta-learning on number_guessing (3 latents, N=10). Part of the [LatentGym](https://huggingface.co/collections/namkoong-lab/latentgym) testbed.
|
|
|
|
## Environments & training latents
|
|
|
|
| Env | Latents seen during training |
|
|
|---|---|
|
|
| number_guessing | `set_of_2`, `set_of_3`, `dynamic_range` |
|
|
|
|
## Training hyperparameters
|
|
|
|
| | |
|
|
|---|---|
|
|
| Base model | `Qwen/Qwen3-8B` |
|
|
| Algorithm | GRPO |
|
|
| Optimizer | AdamW (β₁=0.9, β₂=0.999) |
|
|
| Learning rate | 5e-07 |
|
|
| LR schedule | constant_with_warmup |
|
|
| Weight decay | 0.01 |
|
|
| Max grad norm | 1.0 |
|
|
| KL coefficient β | 0.04 |
|
|
| Clip range ε | 0.2 |
|
|
| Train batch (prompts) | 32 |
|
|
| Mini-batch | 1 |
|
|
| Rollouts per prompt | 8 |
|
|
| Episodes per trajectory (N) | 10 |
|
|
| Reward Φ | Σᵢ Gᵢ (cumulative) |
|
|
| Epochs | 20 |
|
|
| Max generation length | 64 |
|
|
| Sampling (train) | T=0.8, top-p=0.95 |
|
|
| Seed | 263 |
|