|
|
| --- |
| license: apache-2.0 |
| dataset: sft |
| tags: |
| - finetuned |
| - multimodal |
| inference: false |
| --- |
| |
| These are weights for a version of `checkpoints/stage2/llava-moleculestm-vicuna-7b-v1.5-pretrain_rxn_nc` finetuned for multimodal applications. |
|
|
| ### Modalities |
|
|
| * Molecule2DModality (use `<molecule_2d>` in text and provide `molecules` |
|
|
| ### Usage |
|
|
| GitHub: https://github.com/IDEA-XL/PRESTO (includes training scripts and basic inference server) |
|
|
| ### Dataset |
|
|
| sft (765299 examples) |
|
|
| ### Training Device(s) |
|
|
| ``` |
| name, pci.bus_id, vbios_version |
| A100-SXM4-40GB, 00000000:07:00.0, 92.00.45.00.03 |
| A100-SXM4-40GB, 00000000:0F:00.0, 92.00.45.00.03 |
| A100-SXM4-40GB, 00000000:47:00.0, 92.00.45.00.03 |
| A100-SXM4-40GB, 00000000:4E:00.0, 92.00.45.00.03 |
| A100-SXM4-40GB, 00000000:87:00.0, 92.00.45.00.03 |
| A100-SXM4-40GB, 00000000:90:00.0, 92.00.45.00.03 |
| A100-SXM4-40GB, 00000000:B7:00.0, 92.00.45.00.03 |
| A100-SXM4-40GB, 00000000:BD:00.0, 92.00.45.00.03 |
| ``` |
|
|
|
|
| ### Model |
|
|
| ``` |
| LlamaLMMForCausalLM.model = |
| |
| LlamaLMMForCausalLM( |
| (model): LlamaLMMModel( |
| (embed_tokens): Embedding(32000, 4096, padding_idx=0) |
| (layers): ModuleList( |
| (0-31): 32 x LlamaDecoderLayer( |
| (self_attn): LlamaSdpaAttention( |
| (q_proj): Linear(in_features=4096, out_features=4096, bias=False) |
| (k_proj): Linear(in_features=4096, out_features=4096, bias=False) |
| (v_proj): Linear(in_features=4096, out_features=4096, bias=False) |
| (o_proj): Linear(in_features=4096, out_features=4096, bias=False) |
| (rotary_emb): LlamaRotaryEmbedding() |
| ) |
| (mlp): LlamaMLP( |
| (gate_proj): Linear(in_features=4096, out_features=11008, bias=False) |
| (up_proj): Linear(in_features=4096, out_features=11008, bias=False) |
| (down_proj): Linear(in_features=11008, out_features=4096, bias=False) |
| (act_fn): SiLU() |
| ) |
| (input_layernorm): LlamaRMSNorm() |
| (post_attention_layernorm): LlamaRMSNorm() |
| ) |
| ) |
| (norm): LlamaRMSNorm() |
| (molecule_2d_lmm_projector): _MLPVectorProjector( |
| (mlp): Sequential( |
| (0): Linear(in_features=300, out_features=4096, bias=True) |
| (1): GELU(approximate='none') |
| (2): Linear(in_features=4096, out_features=4096, bias=True) |
| ) |
| ) |
| ) |
| (lm_head): Linear(in_features=4096, out_features=32000, bias=False) |
| ) |
| ``` |
|
|
|
|
