Training in progress, step 13000
Browse files- adapter_model.safetensors +1 -1
- config.json +27 -21
- model.safetensors +2 -2
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 11301520
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0a15c640269b10d8e33856a5ca0dd6612cd37d02f1b45a934b5c542f341afafb
|
| 3 |
size 11301520
|
config.json
CHANGED
|
@@ -1,31 +1,37 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"architectures": [
|
| 4 |
-
"
|
| 5 |
],
|
|
|
|
| 6 |
"attention_dropout": 0.0,
|
| 7 |
-
"bos_token_id":
|
| 8 |
-
"eos_token_id":
|
|
|
|
| 9 |
"hidden_act": "silu",
|
| 10 |
-
"hidden_size":
|
| 11 |
"initializer_range": 0.02,
|
| 12 |
-
"intermediate_size":
|
| 13 |
-
"max_position_embeddings":
|
| 14 |
-
"
|
| 15 |
-
"model_type": "
|
| 16 |
-
"num_attention_heads":
|
| 17 |
-
"num_hidden_layers":
|
| 18 |
-
"num_key_value_heads":
|
| 19 |
-
"pad_token_id":
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"
|
| 23 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
"tie_word_embeddings": true,
|
| 25 |
-
"torch_dtype": "
|
| 26 |
"transformers_version": "4.48.2",
|
| 27 |
"use_cache": true,
|
| 28 |
-
"
|
| 29 |
-
"use_sliding_window": false,
|
| 30 |
-
"vocab_size": 151936
|
| 31 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "llama_fourier_cnt_pretrain",
|
| 3 |
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 128000,
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_scaling": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_type": "llama3"
|
| 30 |
+
},
|
| 31 |
+
"rope_theta": 500000.0,
|
| 32 |
"tie_word_embeddings": true,
|
| 33 |
+
"torch_dtype": "float32",
|
| 34 |
"transformers_version": "4.48.2",
|
| 35 |
"use_cache": true,
|
| 36 |
+
"vocab_size": 128256
|
|
|
|
|
|
|
| 37 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:626f785dc5e80339a672f2a321c71374d44b3f08630fa35f4cf923d0ba4f9a80
|
| 3 |
+
size 4943274328
|