Qwen3.5-2B-Translation-TogetherAI / adapter_config.json
FormlessAI's picture
Upload folder using huggingface_hub
330c97a verified
Raw
History Blame Contribute Delete
10.3 kB
{
"alpha_pattern": {},
"auto_mapping": null,
"base_model_name_or_path": "togethercomputer/Qwen3.5-2B",
"bias": "none",
"corda_config": null,
"eva_config": null,
"exclude_modules": [
"model.visual.blocks.8.attn.proj",
"model.visual.blocks.22.mlp.linear_fc1",
"model.visual.blocks.17.norm2",
"model.visual.blocks.12.attn.qkv",
"model.visual.blocks.17.attn.proj",
"model.visual.blocks.19.attn.qkv",
"model.visual.blocks.19.mlp",
"model.visual.blocks.21.attn.proj",
"model.visual.blocks.0.norm1",
"model.visual.blocks.11.mlp.linear_fc2",
"model.visual.blocks.17.norm1",
"model.visual.blocks.6.attn",
"model.visual.blocks.5.mlp.linear_fc1",
"model.visual.blocks.0.norm2",
"model.visual.blocks.11.norm1",
"model.visual.blocks.13.attn",
"model.visual.blocks.16.attn.proj",
"model.visual.blocks.17.mlp",
"model.visual.blocks.19.attn",
"model.visual.blocks.20",
"model.visual.blocks.2.norm1",
"model.visual.blocks.15.mlp.linear_fc2",
"model.visual.blocks.20.attn",
"model.visual.blocks.4.attn.qkv",
"model.visual.blocks.10.attn.proj",
"model.visual.blocks.16",
"model.visual.blocks.12.mlp.act_fn",
"model.visual.blocks.3.mlp",
"model.visual.patch_embed.proj",
"model.visual.blocks.21.attn.qkv",
"model.visual.blocks.17.mlp.linear_fc2",
"model.visual.blocks.3.mlp.linear_fc1",
"model.visual.blocks.9.attn.proj",
"model.visual.blocks.19.mlp.linear_fc1",
"model.visual.blocks.2.mlp.linear_fc2",
"model.visual.blocks.14.norm1",
"model.visual.blocks.0.mlp",
"model.visual.blocks.20.mlp.act_fn",
"model.visual.blocks.16.mlp",
"model.visual.blocks.23.mlp",
"model.visual.blocks.7.mlp",
"model.visual.blocks.9.attn",
"model.visual.blocks.20.attn.qkv",
"model.visual.blocks.5.mlp.linear_fc2",
"model.visual.blocks.22.attn",
"model.visual.blocks.11.norm2",
"model.visual.blocks.14.mlp.linear_fc2",
"model.visual.blocks.6.attn.proj",
"model.visual.blocks.1.mlp.linear_fc2",
"model.visual.pos_embed",
"model.visual.blocks.1.norm2",
"model.visual.blocks.18.norm2",
"model.visual.blocks",
"model.visual.blocks.7.mlp.linear_fc2",
"model.visual.blocks.4.mlp.linear_fc1",
"model.visual.blocks.23",
"model.visual.blocks.9.attn.qkv",
"model.visual.blocks.13.mlp",
"model.visual.blocks.15.mlp.linear_fc1",
"model.visual.blocks.13.attn.qkv",
"model.visual.blocks.21.attn",
"model.visual.blocks.0.attn.proj",
"model.visual.blocks.0.mlp.linear_fc2",
"model.visual.blocks.19.mlp.linear_fc2",
"model.visual.blocks.22.norm2",
"model.visual.blocks.1",
"model.visual.blocks.18.mlp",
"model.visual.blocks.1.attn",
"model.visual.blocks.1.attn.proj",
"model.visual.blocks.12.mlp.linear_fc1",
"model.visual.blocks.5.mlp",
"model.visual.blocks.10.attn.qkv",
"model.visual.blocks.14.attn",
"model.visual.blocks.22.attn.proj",
"model.visual.blocks.23.attn.proj",
"model.visual.blocks.3.attn.proj",
"model.visual.blocks.14.mlp.linear_fc1",
"model.visual.blocks.15.attn",
"model.visual.blocks.23.mlp.linear_fc2",
"model.visual.blocks.8.attn",
"model.visual.blocks.5.norm2",
"model.visual.merger.linear_fc2",
"model.visual.blocks.12.mlp",
"model.visual.blocks.14",
"model.visual.blocks.10.mlp.linear_fc1",
"model.visual.blocks.15.attn.proj",
"model.visual.blocks.19.mlp.act_fn",
"model.visual.blocks.5.attn.proj",
"model.visual.merger.linear_fc1",
"model.visual.blocks.11.attn.qkv",
"model.visual.blocks.12",
"model.visual.blocks.3.mlp.act_fn",
"model.visual.blocks.17",
"model.visual.blocks.9",
"model.visual.blocks.17.attn",
"model.visual.blocks.3.attn",
"model.visual.blocks.1.mlp",
"model.visual.blocks.22.mlp.linear_fc2",
"model.visual.blocks.22",
"model.visual.blocks.23.mlp.act_fn",
"model.visual.blocks.6.mlp.linear_fc2",
"model.visual.blocks.23.norm2",
"model.visual.blocks.1.attn.qkv",
"model.visual.blocks.10.norm2",
"model.visual.blocks.5.norm1",
"model.visual.blocks.0.mlp.act_fn",
"model.visual.blocks.9.norm2",
"model.visual.blocks.18",
"model.visual.blocks.21.mlp",
"model.visual.blocks.11",
"model.visual.blocks.7.attn",
"model.visual.blocks.5.attn",
"model.visual.blocks.8.mlp.linear_fc1",
"model.visual.blocks.13.mlp.act_fn",
"model.visual.blocks.22.attn.qkv",
"model.visual.blocks.21.norm1",
"model.visual.blocks.23.attn",
"model.visual.blocks.6.mlp.act_fn",
"model.visual.blocks.10.attn",
"model.visual.blocks.7.mlp.linear_fc1",
"model.visual.blocks.2.mlp",
"model.visual.blocks.21",
"model.visual.blocks.1.mlp.linear_fc1",
"model.visual.blocks.13.mlp.linear_fc2",
"model.visual.blocks.6.mlp",
"model.visual.blocks.19",
"model.visual.blocks.22.mlp",
"model.visual.blocks.10.mlp.linear_fc2",
"model.visual.blocks.5",
"model.visual.blocks.6",
"model.visual.blocks.17.mlp.act_fn",
"model.visual.blocks.13.attn.proj",
"model.visual.merger.norm",
"model.visual.blocks.22.mlp.act_fn",
"model.visual.blocks.10.mlp.act_fn",
"model.visual.blocks.20.mlp.linear_fc1",
"model.visual.blocks.11.mlp.act_fn",
"model.visual.blocks.8.norm2",
"model.visual.blocks.11.attn",
"model.visual.blocks.15.norm2",
"model.visual.merger.act_fn",
"model.visual.blocks.4.norm2",
"model.visual.blocks.16.attn",
"model.visual.blocks.21.mlp.linear_fc1",
"model.visual.blocks.16.mlp.act_fn",
"model.visual.blocks.0",
"model.visual.blocks.3.norm1",
"model.visual.blocks.2.attn",
"model.visual.blocks.20.norm2",
"model.visual.blocks.7.attn.proj",
"model.visual.blocks.4.attn.proj",
"model.visual.blocks.19.norm2",
"model.visual.blocks.7",
"model.visual.blocks.11.mlp",
"model.visual.blocks.17.attn.qkv",
"model.visual.blocks.8.mlp.act_fn",
"model.visual.blocks.18.mlp.linear_fc2",
"model.visual.blocks.1.norm1",
"model.visual.blocks.3",
"model.visual.blocks.0.mlp.linear_fc1",
"model.visual.blocks.5.attn.qkv",
"model.visual.blocks.15.mlp",
"model.visual.blocks.16.norm2",
"model.visual.blocks.14.attn.proj",
"model.visual.blocks.3.mlp.linear_fc2",
"model.visual.blocks.13.norm2",
"model.visual.blocks.18.mlp.act_fn",
"model.visual.blocks.10.mlp",
"model.visual.blocks.13.norm1",
"model.visual.blocks.11.mlp.linear_fc1",
"model.visual.blocks.12.mlp.linear_fc2",
"model.visual.blocks.18.attn",
"model.visual.blocks.10",
"model.visual.blocks.2.attn.proj",
"model.visual",
"model.visual.blocks.11.attn.proj",
"model.visual.blocks.17.mlp.linear_fc1",
"model.visual.blocks.18.attn.proj",
"model.visual.blocks.7.mlp.act_fn",
"model.visual.blocks.16.mlp.linear_fc2",
"model.visual.blocks.14.norm2",
"model.visual.blocks.19.norm1",
"model.visual.blocks.4.mlp",
"model.visual.blocks.6.norm1",
"model.visual.blocks.4.attn",
"model.visual.blocks.8",
"model.visual.patch_embed",
"model.visual.blocks.6.norm2",
"model.visual.blocks.10.norm1",
"model.visual.blocks.4.mlp.act_fn",
"model.visual.blocks.4.norm1",
"model.visual.blocks.8.mlp",
"model.visual.blocks.20.mlp",
"model.visual.blocks.1.mlp.act_fn",
"model.visual.blocks.3.attn.qkv",
"model.visual.blocks.8.mlp.linear_fc2",
"model.visual.blocks.0.attn.qkv",
"model.visual.blocks.2",
"model.visual.blocks.22.norm1",
"model.visual.blocks.7.attn.qkv",
"model.visual.blocks.18.mlp.linear_fc1",
"model.visual.blocks.7.norm1",
"model.visual.blocks.13",
"model.visual.blocks.16.norm1",
"model.visual.blocks.21.mlp.linear_fc2",
"model.visual.blocks.23.mlp.linear_fc1",
"model.visual.blocks.15.mlp.act_fn",
"model.visual.blocks.9.mlp",
"model.visual.blocks.2.mlp.act_fn",
"model.visual.blocks.6.mlp.linear_fc1",
"model.visual.blocks.20.attn.proj",
"model.visual.blocks.21.mlp.act_fn",
"model.visual.blocks.2.norm2",
"model.visual.blocks.8.norm1",
"model.visual.blocks.12.norm1",
"model.visual.blocks.2.mlp.linear_fc1",
"model.visual.blocks.4",
"model.visual.blocks.23.attn.qkv",
"model.visual.blocks.12.norm2",
"model.visual.blocks.9.mlp.linear_fc2",
"model.visual.blocks.13.mlp.linear_fc1",
"model.visual.blocks.3.norm2",
"model.visual.blocks.9.mlp.act_fn",
"model.visual.blocks.16.mlp.linear_fc1",
"model.visual.blocks.14.mlp.act_fn",
"model.visual.blocks.4.mlp.linear_fc2",
"model.visual.blocks.18.attn.qkv",
"model.visual.blocks.12.attn",
"model.visual.blocks.19.attn.proj",
"model.visual.blocks.21.norm2",
"model.visual.merger",
"model.visual.blocks.20.mlp.linear_fc2",
"model.visual.blocks.7.norm2",
"model.visual.blocks.9.norm1",
"model.visual.blocks.6.attn.qkv",
"model.visual.blocks.14.attn.qkv",
"model.visual.blocks.14.mlp",
"model.visual.blocks.23.norm1",
"model.visual.blocks.12.attn.proj",
"model.visual.blocks.0.attn",
"model.visual.blocks.15.norm1",
"model.visual.blocks.2.attn.qkv",
"model.visual.blocks.15.attn.qkv",
"model.visual.blocks.20.norm1",
"model.visual.blocks.15",
"model.visual.blocks.8.attn.qkv",
"model.visual.rotary_pos_emb",
"model.visual.blocks.18.norm1",
"model.visual.blocks.16.attn.qkv",
"model.visual.blocks.5.mlp.act_fn",
"model.visual.blocks.9.mlp.linear_fc1"
],
"fan_in_fan_out": false,
"inference_mode": true,
"init_lora_weights": true,
"layer_replication": null,
"layers_pattern": null,
"layers_to_transform": null,
"loftq_config": {},
"lora_alpha": 128,
"lora_bias": false,
"lora_dropout": 0.0,
"megatron_config": null,
"megatron_core": "megatron.core",
"modules_to_save": null,
"peft_type": "LORA",
"r": 64,
"rank_pattern": {},
"revision": null,
"target_modules": [
"up_proj",
"o_proj",
"k_proj",
"down_proj",
"gate_proj",
"v_proj",
"q_proj"
],
"task_type": "CAUSAL_LM",
"trainable_token_indices": null,
"use_dora": false,
"use_rslora": false
}