Feature Extraction
Transformers
Safetensors
llama
financial
transactions
foundation-model
embeddings
Instructions to use simonykq/nvidia-transaction-decoder-fm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use simonykq/nvidia-transaction-decoder-fm with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="simonykq/nvidia-transaction-decoder-fm")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("simonykq/nvidia-transaction-decoder-fm", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload folder using huggingface_hub
Browse files- README.md +22 -0
- config.json +27 -0
- generation_config.json +7 -0
- model-00001-of-00001.safetensors +3 -0
- model.safetensors.index.json +82 -0
README.md
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: transformers
|
| 4 |
+
tags:
|
| 5 |
+
- financial
|
| 6 |
+
- transactions
|
| 7 |
+
- llama
|
| 8 |
+
- foundation-model
|
| 9 |
+
- embeddings
|
| 10 |
+
pipeline_tag: feature-extraction
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# NVIDIA Transaction Decoder Foundation Model (~29M)
|
| 14 |
+
|
| 15 |
+
Pretrained checkpoint from the [NVIDIA Build Your Own Transaction Foundation Model](https://github.com/NVIDIA-AI-Blueprints/transaction-foundation-model) developer example, as described in the [NVIDIA Technical Blog](https://developer.nvidia.com/blog/build-your-own-transaction-foundation-model-for-financial-intelligence/).
|
| 16 |
+
|
| 17 |
+
- Architecture: Llama decoder-only (~29M params, hidden 512, 8 layers, GQA)
|
| 18 |
+
- Vocabulary: ~6,251 domain-specific financial transaction tokens
|
| 19 |
+
- Trained ~3,000 steps on IBM TabFormer with NeMo AutoModel
|
| 20 |
+
- Load with `AutoModelForCausalLM.from_pretrained(...)` and extract last-token embeddings
|
| 21 |
+
|
| 22 |
+
This Hub mirror exists so Spaces / Transformers can load the Git LFS checkpoint published under `models/decoder-foundation-model/` in the upstream GitHub repo.
|
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"attention_bias": false,
|
| 3 |
+
"attention_dropout": 0.0,
|
| 4 |
+
"bos_token_id": 1,
|
| 5 |
+
"eos_token_id": 2,
|
| 6 |
+
"head_dim": 64,
|
| 7 |
+
"hidden_act": "silu",
|
| 8 |
+
"hidden_size": 512,
|
| 9 |
+
"initializer_range": 0.02,
|
| 10 |
+
"intermediate_size": 1408,
|
| 11 |
+
"max_position_embeddings": 8192,
|
| 12 |
+
"mlp_bias": false,
|
| 13 |
+
"model_type": "llama",
|
| 14 |
+
"nemo_version": "0.1.1",
|
| 15 |
+
"num_attention_heads": 8,
|
| 16 |
+
"num_hidden_layers": 8,
|
| 17 |
+
"num_key_value_heads": 2,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"pretraining_tp": 1,
|
| 20 |
+
"rms_norm_eps": 1e-05,
|
| 21 |
+
"rope_scaling": null,
|
| 22 |
+
"rope_theta": 500000.0,
|
| 23 |
+
"tie_word_embeddings": false,
|
| 24 |
+
"transformers_version": "4.53.3",
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 6251
|
| 27 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.53.3"
|
| 7 |
+
}
|
model-00001-of-00001.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:71c5899c2ab0146c209207d67efc65dda882f4966ae09877b2580c14bcb62cbe
|
| 3 |
+
size 57917064
|
model.safetensors.index.json
ADDED
|
@@ -0,0 +1,82 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"metadata": {
|
| 3 |
+
"total_size": 57908224
|
| 4 |
+
},
|
| 5 |
+
"weight_map": {
|
| 6 |
+
"model.embed_tokens.weight": "model-00001-of-00001.safetensors",
|
| 7 |
+
"model.layers.0.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 8 |
+
"model.layers.0.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 9 |
+
"model.layers.0.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 10 |
+
"model.layers.0.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 11 |
+
"model.layers.0.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 12 |
+
"model.layers.0.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 13 |
+
"model.layers.0.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 14 |
+
"model.layers.0.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 15 |
+
"model.layers.0.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 16 |
+
"model.layers.1.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 17 |
+
"model.layers.1.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 18 |
+
"model.layers.1.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 19 |
+
"model.layers.1.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 20 |
+
"model.layers.1.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 21 |
+
"model.layers.1.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 22 |
+
"model.layers.1.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 23 |
+
"model.layers.1.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 24 |
+
"model.layers.1.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 25 |
+
"model.layers.2.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 26 |
+
"model.layers.2.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 27 |
+
"model.layers.2.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 28 |
+
"model.layers.2.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 29 |
+
"model.layers.2.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 30 |
+
"model.layers.2.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 31 |
+
"model.layers.2.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 32 |
+
"model.layers.2.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 33 |
+
"model.layers.2.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 34 |
+
"model.layers.3.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 35 |
+
"model.layers.3.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 36 |
+
"model.layers.3.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 37 |
+
"model.layers.3.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 38 |
+
"model.layers.3.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 39 |
+
"model.layers.3.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 40 |
+
"model.layers.3.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 41 |
+
"model.layers.3.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 42 |
+
"model.layers.3.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 43 |
+
"model.layers.4.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 44 |
+
"model.layers.4.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 45 |
+
"model.layers.4.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 46 |
+
"model.layers.4.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 47 |
+
"model.layers.4.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 48 |
+
"model.layers.4.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 49 |
+
"model.layers.4.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 50 |
+
"model.layers.4.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 51 |
+
"model.layers.4.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 52 |
+
"model.layers.5.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 53 |
+
"model.layers.5.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 54 |
+
"model.layers.5.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 55 |
+
"model.layers.5.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 56 |
+
"model.layers.5.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 57 |
+
"model.layers.5.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 58 |
+
"model.layers.5.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 59 |
+
"model.layers.5.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 60 |
+
"model.layers.5.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 61 |
+
"model.layers.6.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 62 |
+
"model.layers.6.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 63 |
+
"model.layers.6.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 64 |
+
"model.layers.6.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 65 |
+
"model.layers.6.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 66 |
+
"model.layers.6.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 67 |
+
"model.layers.6.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 68 |
+
"model.layers.6.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 69 |
+
"model.layers.6.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 70 |
+
"model.layers.7.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
|
| 71 |
+
"model.layers.7.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
|
| 72 |
+
"model.layers.7.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
|
| 73 |
+
"model.layers.7.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
|
| 74 |
+
"model.layers.7.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
|
| 75 |
+
"model.layers.7.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
|
| 76 |
+
"model.layers.7.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
|
| 77 |
+
"model.layers.7.input_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 78 |
+
"model.layers.7.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
|
| 79 |
+
"model.norm.weight": "model-00001-of-00001.safetensors",
|
| 80 |
+
"lm_head.weight": "model-00001-of-00001.safetensors"
|
| 81 |
+
}
|
| 82 |
+
}
|