Update adapter_config.json
Browse files- adapter_config.json +32 -32
adapter_config.json
CHANGED
@@ -1,35 +1,35 @@
|
|
1 |
{
|
2 |
-
"
|
3 |
-
|
4 |
-
"base_model_name_or_path": "unsloth/gemma-2-2b-bnb-4bit",
|
5 |
-
"bias": "none",
|
6 |
-
"fan_in_fan_out": false,
|
7 |
-
"inference_mode": true,
|
8 |
-
"init_lora_weights": true,
|
9 |
-
"layer_replication": null,
|
10 |
-
"layers_pattern": null,
|
11 |
-
"layers_to_transform": null,
|
12 |
-
"loftq_config": {},
|
13 |
-
"lora_alpha": 16,
|
14 |
-
"lora_dropout": 0,
|
15 |
-
"megatron_config": null,
|
16 |
-
"megatron_core": "megatron.core",
|
17 |
-
"modules_to_save": null,
|
18 |
-
"peft_type": "LORA",
|
19 |
-
"r": 16,
|
20 |
-
"rank_pattern": {},
|
21 |
-
"revision": null,
|
22 |
-
"target_modules": [
|
23 |
-
"k_proj",
|
24 |
-
"v_proj",
|
25 |
-
"gate_proj",
|
26 |
-
"o_proj",
|
27 |
-
"up_proj",
|
28 |
-
"down_proj",
|
29 |
-
"q_proj"
|
30 |
],
|
31 |
-
"
|
32 |
-
"
|
33 |
-
"
|
34 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
35 |
}
|
|
|
1 |
{
|
2 |
+
"architectures": [
|
3 |
+
"Gemma2ForCausalLM"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
],
|
5 |
+
"attention_bias": false,
|
6 |
+
"attention_dropout": 0.0,
|
7 |
+
"attn_logit_softcapping": 50.0,
|
8 |
+
"bos_token_id": 2,
|
9 |
+
"cache_implementation": "hybrid",
|
10 |
+
"eos_token_id": [
|
11 |
+
1,
|
12 |
+
107
|
13 |
+
],
|
14 |
+
"final_logit_softcapping": 30.0,
|
15 |
+
"head_dim": 256,
|
16 |
+
"hidden_act": "gelu_pytorch_tanh",
|
17 |
+
"hidden_activation": "gelu_pytorch_tanh",
|
18 |
+
"hidden_size": 2304,
|
19 |
+
"initializer_range": 0.02,
|
20 |
+
"intermediate_size": 9216,
|
21 |
+
"max_position_embeddings": 8192,
|
22 |
+
"model_type": "gemma2",
|
23 |
+
"num_attention_heads": 8,
|
24 |
+
"num_hidden_layers": 26,
|
25 |
+
"num_key_value_heads": 4,
|
26 |
+
"pad_token_id": 0,
|
27 |
+
"query_pre_attn_scalar": 256,
|
28 |
+
"rms_norm_eps": 1e-06,
|
29 |
+
"rope_theta": 10000.0,
|
30 |
+
"sliding_window": 4096,
|
31 |
+
"torch_dtype": "bfloat16",
|
32 |
+
"transformers_version": "4.42.4",
|
33 |
+
"use_cache": true,
|
34 |
+
"vocab_size": 256000
|
35 |
}
|