Karitasu commited on
Commit
423a3b5
·
verified ·
1 Parent(s): 346af9d

Rename model card to StableLex

Browse files
Files changed (1) hide show
  1. README.md +9 -6
README.md CHANGED
@@ -6,6 +6,7 @@ pipeline_tag: text-generation
6
  language:
7
  - zh
8
  tags:
 
9
  - peft
10
  - lora
11
  - qwen
@@ -15,17 +16,17 @@ tags:
15
  - compliance
16
  ---
17
 
18
- # Qwen3.5 Stablecoin Round3 LoRA Adapter
19
 
20
- This repository provides a PEFT/LoRA adapter for `Qwen/Qwen3.5-9B`. It is not a standalone language model: it must be loaded together with the base model.
21
 
22
- The adapter is intended for controlled stablecoin and virtual-asset analysis workflows where responses are expected to stay grounded in supplied materials, use explicit citations, distinguish source-backed conclusions from cautious inference, and refuse or qualify answers when the provided materials are insufficient.
23
 
24
  ## Repository Contents
25
 
26
  - `adapter_model.safetensors`: LoRA adapter weights.
27
  - `adapter_config.json`: PEFT adapter configuration.
28
- - `formal_eval_report.md`: copied formal evaluation report for the released Round 3 adapter.
29
  - `README.md`: this model card.
30
  - `LICENSE`: Apache License 2.0 for this adapter repository.
31
 
@@ -33,6 +34,9 @@ The adapter is intended for controlled stablecoin and virtual-asset analysis wor
33
 
34
  | Item | Value |
35
  |---|---|
 
 
 
36
  | Base model | `Qwen/Qwen3.5-9B` |
37
  | Adapter type | LoRA via PEFT |
38
  | Task type | `CAUSAL_LM` |
@@ -41,7 +45,6 @@ The adapter is intended for controlled stablecoin and virtual-asset analysis wor
41
  | LoRA alpha | `32` |
42
  | LoRA dropout | `0.05` |
43
  | Target modules | `q_proj`, `k_proj`, `v_proj`, `o_proj`, `gate_proj`, `up_proj`, `down_proj` |
44
- | Source checkpoint | `stablecoin_phase1_pipeline/training/outputs/round3_conservative_repair/checkpoints/full_train_20260602T101206Z/adapter` |
45
  | License | Apache License 2.0 for this adapter repository |
46
 
47
  ## Intended Use
@@ -102,7 +105,7 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
102
  from peft import PeftModel
103
 
104
  base_model_id = "Qwen/Qwen3.5-9B"
105
- adapter_id = "Karitasu/qwen35-stablecoin-round3-lora"
106
 
107
  tokenizer = AutoTokenizer.from_pretrained(
108
  base_model_id,
 
6
  language:
7
  - zh
8
  tags:
9
+ - stablelex
10
  - peft
11
  - lora
12
  - qwen
 
16
  - compliance
17
  ---
18
 
19
+ # StableLex
20
 
21
+ StableLex is a PEFT/LoRA adapter based on `Qwen/Qwen3.5-9B`. It is not a standalone language model: it must be loaded together with the base model.
22
 
23
+ The adapter is intended for controlled stablecoin and virtual-asset legal/compliance analysis workflows where responses are expected to stay grounded in supplied materials, use explicit citations, distinguish source-backed conclusions from cautious inference, and refuse or qualify answers when the provided materials are insufficient.
24
 
25
  ## Repository Contents
26
 
27
  - `adapter_model.safetensors`: LoRA adapter weights.
28
  - `adapter_config.json`: PEFT adapter configuration.
29
+ - `formal_eval_report.md`: copied formal evaluation report for this adapter release.
30
  - `README.md`: this model card.
31
  - `LICENSE`: Apache License 2.0 for this adapter repository.
32
 
 
34
 
35
  | Item | Value |
36
  |---|---|
37
+ | Model name | StableLex |
38
+ | Repository | `Karitasu/StableLex` |
39
+ | Hub URL | `https://huggingface.co/Karitasu/StableLex` |
40
  | Base model | `Qwen/Qwen3.5-9B` |
41
  | Adapter type | LoRA via PEFT |
42
  | Task type | `CAUSAL_LM` |
 
45
  | LoRA alpha | `32` |
46
  | LoRA dropout | `0.05` |
47
  | Target modules | `q_proj`, `k_proj`, `v_proj`, `o_proj`, `gate_proj`, `up_proj`, `down_proj` |
 
48
  | License | Apache License 2.0 for this adapter repository |
49
 
50
  ## Intended Use
 
105
  from peft import PeftModel
106
 
107
  base_model_id = "Qwen/Qwen3.5-9B"
108
+ adapter_id = "Karitasu/StableLex"
109
 
110
  tokenizer = AutoTokenizer.from_pretrained(
111
  base_model_id,