bad output ,try transfer a new
Browse files
text_encoder/config.json
DELETED
|
@@ -1,128 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"Gemma3ForConditionalGeneration"
|
| 4 |
-
],
|
| 5 |
-
"boi_token_index": 255999,
|
| 6 |
-
"dtype": "float16",
|
| 7 |
-
"eoi_token_index": 256000,
|
| 8 |
-
"eos_token_id": [
|
| 9 |
-
1,
|
| 10 |
-
106
|
| 11 |
-
],
|
| 12 |
-
"image_token_index": 262144,
|
| 13 |
-
"initializer_range": 0.02,
|
| 14 |
-
"mm_tokens_per_image": 256,
|
| 15 |
-
"model_type": "gemma3",
|
| 16 |
-
"quantization_config": {
|
| 17 |
-
"_load_in_4bit": true,
|
| 18 |
-
"_load_in_8bit": false,
|
| 19 |
-
"bnb_4bit_compute_dtype": "float32",
|
| 20 |
-
"bnb_4bit_quant_storage": "uint8",
|
| 21 |
-
"bnb_4bit_quant_type": "fp4",
|
| 22 |
-
"bnb_4bit_use_double_quant": false,
|
| 23 |
-
"llm_int8_enable_fp32_cpu_offload": false,
|
| 24 |
-
"llm_int8_has_fp16_weight": false,
|
| 25 |
-
"llm_int8_skip_modules": null,
|
| 26 |
-
"llm_int8_threshold": 6.0,
|
| 27 |
-
"load_in_4bit": true,
|
| 28 |
-
"load_in_8bit": false,
|
| 29 |
-
"quant_method": "bitsandbytes"
|
| 30 |
-
},
|
| 31 |
-
"text_config": {
|
| 32 |
-
"_sliding_window_pattern": 6,
|
| 33 |
-
"attention_bias": false,
|
| 34 |
-
"attention_dropout": 0.0,
|
| 35 |
-
"attn_logit_softcapping": null,
|
| 36 |
-
"cache_implementation": "hybrid",
|
| 37 |
-
"dtype": "float16",
|
| 38 |
-
"final_logit_softcapping": null,
|
| 39 |
-
"head_dim": 256,
|
| 40 |
-
"hidden_activation": "gelu_pytorch_tanh",
|
| 41 |
-
"hidden_size": 3840,
|
| 42 |
-
"initializer_range": 0.02,
|
| 43 |
-
"intermediate_size": 15360,
|
| 44 |
-
"layer_types": [
|
| 45 |
-
"sliding_attention",
|
| 46 |
-
"sliding_attention",
|
| 47 |
-
"sliding_attention",
|
| 48 |
-
"sliding_attention",
|
| 49 |
-
"sliding_attention",
|
| 50 |
-
"full_attention",
|
| 51 |
-
"sliding_attention",
|
| 52 |
-
"sliding_attention",
|
| 53 |
-
"sliding_attention",
|
| 54 |
-
"sliding_attention",
|
| 55 |
-
"sliding_attention",
|
| 56 |
-
"full_attention",
|
| 57 |
-
"sliding_attention",
|
| 58 |
-
"sliding_attention",
|
| 59 |
-
"sliding_attention",
|
| 60 |
-
"sliding_attention",
|
| 61 |
-
"sliding_attention",
|
| 62 |
-
"full_attention",
|
| 63 |
-
"sliding_attention",
|
| 64 |
-
"sliding_attention",
|
| 65 |
-
"sliding_attention",
|
| 66 |
-
"sliding_attention",
|
| 67 |
-
"sliding_attention",
|
| 68 |
-
"full_attention",
|
| 69 |
-
"sliding_attention",
|
| 70 |
-
"sliding_attention",
|
| 71 |
-
"sliding_attention",
|
| 72 |
-
"sliding_attention",
|
| 73 |
-
"sliding_attention",
|
| 74 |
-
"full_attention",
|
| 75 |
-
"sliding_attention",
|
| 76 |
-
"sliding_attention",
|
| 77 |
-
"sliding_attention",
|
| 78 |
-
"sliding_attention",
|
| 79 |
-
"sliding_attention",
|
| 80 |
-
"full_attention",
|
| 81 |
-
"sliding_attention",
|
| 82 |
-
"sliding_attention",
|
| 83 |
-
"sliding_attention",
|
| 84 |
-
"sliding_attention",
|
| 85 |
-
"sliding_attention",
|
| 86 |
-
"full_attention",
|
| 87 |
-
"sliding_attention",
|
| 88 |
-
"sliding_attention",
|
| 89 |
-
"sliding_attention",
|
| 90 |
-
"sliding_attention",
|
| 91 |
-
"sliding_attention",
|
| 92 |
-
"full_attention"
|
| 93 |
-
],
|
| 94 |
-
"max_position_embeddings": 131072,
|
| 95 |
-
"model_type": "gemma3_text",
|
| 96 |
-
"num_attention_heads": 16,
|
| 97 |
-
"num_hidden_layers": 48,
|
| 98 |
-
"num_key_value_heads": 8,
|
| 99 |
-
"query_pre_attn_scalar": 256,
|
| 100 |
-
"rms_norm_eps": 1e-06,
|
| 101 |
-
"rope_local_base_freq": 10000,
|
| 102 |
-
"rope_scaling": {
|
| 103 |
-
"factor": 8.0,
|
| 104 |
-
"rope_type": "linear"
|
| 105 |
-
},
|
| 106 |
-
"rope_theta": 1000000,
|
| 107 |
-
"sliding_window": 1024,
|
| 108 |
-
"use_bidirectional_attention": false,
|
| 109 |
-
"use_cache": true,
|
| 110 |
-
"vocab_size": 262208
|
| 111 |
-
},
|
| 112 |
-
"transformers_version": "4.57.3",
|
| 113 |
-
"vision_config": {
|
| 114 |
-
"attention_dropout": 0.0,
|
| 115 |
-
"dtype": "float16",
|
| 116 |
-
"hidden_act": "gelu_pytorch_tanh",
|
| 117 |
-
"hidden_size": 1152,
|
| 118 |
-
"image_size": 896,
|
| 119 |
-
"intermediate_size": 4304,
|
| 120 |
-
"layer_norm_eps": 1e-06,
|
| 121 |
-
"model_type": "siglip_vision_model",
|
| 122 |
-
"num_attention_heads": 16,
|
| 123 |
-
"num_channels": 3,
|
| 124 |
-
"num_hidden_layers": 27,
|
| 125 |
-
"patch_size": 14,
|
| 126 |
-
"vision_use_head": false
|
| 127 |
-
}
|
| 128 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
text_encoder/generation_config.json
DELETED
|
@@ -1,11 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"cache_implementation": "hybrid",
|
| 3 |
-
"do_sample": true,
|
| 4 |
-
"eos_token_id": [
|
| 5 |
-
1,
|
| 6 |
-
106
|
| 7 |
-
],
|
| 8 |
-
"top_k": 64,
|
| 9 |
-
"top_p": 0.95,
|
| 10 |
-
"transformers_version": "4.57.3"
|
| 11 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
text_encoder/model-00001-of-00002.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:f76a94367cd5391d280222c584ec2812fec5f6d6adac8430f2a20e87c0963762
|
| 3 |
-
size 4973662978
|
|
|
|
|
|
|
|
|
|
|
|
text_encoder/model-00002-of-00002.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:87eab8931d28eec0c136e0efacb004d91878b574829d8df011c17afacfc90c7f
|
| 3 |
-
size 3345313504
|
|
|
|
|
|
|
|
|
|
|
|
text_encoder/model.safetensors.index.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|