Upload folder using huggingface_hub
Browse files
opt-125m-llama3-8b-lmsys-class-trainbucket820-b32/finetuned/config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/opt-125m",
|
| 3 |
+
"_remove_final_layer_norm": false,
|
| 4 |
+
"activation_dropout": 0.0,
|
| 5 |
+
"activation_function": "relu",
|
| 6 |
+
"architectures": [
|
| 7 |
+
"OPTForSequenceClassification"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 2,
|
| 11 |
+
"do_layer_norm_before": true,
|
| 12 |
+
"dropout": 0.1,
|
| 13 |
+
"enable_bias": true,
|
| 14 |
+
"eos_token_id": 2,
|
| 15 |
+
"ffn_dim": 3072,
|
| 16 |
+
"hidden_size": 768,
|
| 17 |
+
"id2label": {
|
| 18 |
+
"0": "LABEL_0",
|
| 19 |
+
"1": "LABEL_1",
|
| 20 |
+
"2": "LABEL_2",
|
| 21 |
+
"3": "LABEL_3",
|
| 22 |
+
"4": "LABEL_4",
|
| 23 |
+
"5": "LABEL_5",
|
| 24 |
+
"6": "LABEL_6",
|
| 25 |
+
"7": "LABEL_7",
|
| 26 |
+
"8": "LABEL_8",
|
| 27 |
+
"9": "LABEL_9"
|
| 28 |
+
},
|
| 29 |
+
"init_std": 0.02,
|
| 30 |
+
"label2id": {
|
| 31 |
+
"LABEL_0": 0,
|
| 32 |
+
"LABEL_1": 1,
|
| 33 |
+
"LABEL_2": 2,
|
| 34 |
+
"LABEL_3": 3,
|
| 35 |
+
"LABEL_4": 4,
|
| 36 |
+
"LABEL_5": 5,
|
| 37 |
+
"LABEL_6": 6,
|
| 38 |
+
"LABEL_7": 7,
|
| 39 |
+
"LABEL_8": 8,
|
| 40 |
+
"LABEL_9": 9
|
| 41 |
+
},
|
| 42 |
+
"layer_norm_elementwise_affine": true,
|
| 43 |
+
"layerdrop": 0.0,
|
| 44 |
+
"max_position_embeddings": 2048,
|
| 45 |
+
"model_type": "opt",
|
| 46 |
+
"num_attention_heads": 12,
|
| 47 |
+
"num_hidden_layers": 12,
|
| 48 |
+
"num_labels": 10,
|
| 49 |
+
"pad_token_id": 1,
|
| 50 |
+
"prefix": "</s>",
|
| 51 |
+
"torch_dtype": "float16",
|
| 52 |
+
"transformers_version": "4.45.2",
|
| 53 |
+
"use_cache": true,
|
| 54 |
+
"vocab_size": 50272,
|
| 55 |
+
"word_embed_proj_dim": 768
|
| 56 |
+
}
|
opt-125m-llama3-8b-lmsys-class-trainbucket820-b32/finetuned/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ae91e23239037bf18d5a7a4fad904a47f9254becda4a8adf0a5f294dfd888db
|
| 3 |
+
size 250516416
|
opt-125m-llama3-8b-lmsys-class-trainbucket820-b32/usage_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"model": {"pred_model": "facebook/opt-125m", "num_labels": 10, "mtype": "class", "activation": null, "path": "MODEL/results/opt-125m-llama3-8b-lmsys-class-trainbucket820-b32/finetuned", "max_length": 2048, "max_batch_size": 1000}}
|