43 lines
1.1 KiB
Markdown
43 lines
1.1 KiB
Markdown
|
---
|
||
|
|
base_model: Qwen/Qwen2.5-7B-Instruct
|
||
|
|
datasets:
|
||
|
|
- Yano/exp-0223-027-realobs-llmagent-alfworld-data
|
||
|
|
language:
|
||
|
|
- en
|
||
|
|
license: apache-2.0
|
||
|
|
library_name: transformers
|
||
|
|
pipeline_tag: text-generation
|
||
|
|
tags:
|
||
|
|
- qlora
|
||
|
|
- lora
|
||
|
|
- merged
|
||
|
|
- alfworld
|
||
|
|
- agent
|
||
|
|
- realobs-llmagent
|
||
|
|
---
|
||
|
|
|
||
|
|
# exp-0223-027-realobs-llmagent-qwen2.5-7b
|
||
|
|
|
||
|
|
Fine-tuned from **Qwen/Qwen2.5-7B-Instruct** using **QLoRA (4-bit, Unsloth)**.
|
||
|
|
|
||
|
|
## Purpose
|
||
|
|
|
||
|
|
ALFWorld SFT combining real environment observations from experiment 016
|
||
|
|
with LLM-renarrated agent responses (strategic THOUGHTs + ACTION-dominant format).
|
||
|
|
|
||
|
|
## Key Design
|
||
|
|
- Environment responses (user): Real data from 016 (byte-exact preservation)
|
||
|
|
- Agent responses (assistant): LLM-renarrated (strategic THOUGHT, not template)
|
||
|
|
- Format: First turn THOUGHT+ACTION, subsequent ACTION-only, recovery THOUGHT after failures
|
||
|
|
- Failure patterns: Naturally inherited from 016's 72% failure-containing trajectories
|
||
|
|
|
||
|
|
## Training Configuration
|
||
|
|
|
||
|
|
- Base model: Qwen/Qwen2.5-7B-Instruct
|
||
|
|
- Method: QLoRA (4-bit), merged to 16-bit
|
||
|
|
- Max sequence length: 2048
|
||
|
|
- Epochs: 3
|
||
|
|
- Learning rate: 2e-05
|
||
|
|
- LoRA: r=64, alpha=128
|
||
|
|
- Collator: AllAssistantTurnsCollator (all turns supervised)
|