Commit
•
081a23e
1
Parent(s):
c43da3e
(Trained with Unsloth)
Browse files- config.json +2 -34
config.json
CHANGED
|
@@ -1,35 +1,3 @@
|
|
| 1 |
{
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
"Gemma2ForCausalLM"
|
| 5 |
-
],
|
| 6 |
-
"attention_bias": false,
|
| 7 |
-
"attention_dropout": 0.0,
|
| 8 |
-
"attn_logit_softcapping": 50.0,
|
| 9 |
-
"bos_token_id": 2,
|
| 10 |
-
"cache_implementation": "hybrid",
|
| 11 |
-
"eos_token_id": 1,
|
| 12 |
-
"final_logit_softcapping": 30.0,
|
| 13 |
-
"head_dim": 256,
|
| 14 |
-
"hidden_act": "gelu_pytorch_tanh",
|
| 15 |
-
"hidden_activation": "gelu_pytorch_tanh",
|
| 16 |
-
"hidden_size": 2304,
|
| 17 |
-
"initializer_range": 0.02,
|
| 18 |
-
"intermediate_size": 9216,
|
| 19 |
-
"max_position_embeddings": 8192,
|
| 20 |
-
"model_type": "gemma2",
|
| 21 |
-
"num_attention_heads": 8,
|
| 22 |
-
"num_hidden_layers": 26,
|
| 23 |
-
"num_key_value_heads": 4,
|
| 24 |
-
"pad_token_id": 0,
|
| 25 |
-
"query_pre_attn_scalar": 256,
|
| 26 |
-
"rms_norm_eps": 1e-06,
|
| 27 |
-
"rope_scaling": null,
|
| 28 |
-
"rope_theta": 10000.0,
|
| 29 |
-
"sliding_window": 4096,
|
| 30 |
-
"torch_dtype": "bfloat16",
|
| 31 |
-
"transformers_version": "4.43.3",
|
| 32 |
-
"unsloth_version": "2024.8",
|
| 33 |
-
"use_cache": true,
|
| 34 |
-
"vocab_size": 256000
|
| 35 |
-
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"model_type": "gemma2"
|
| 3 |
+
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|