r19v1_dense3 (R19, cutoff 2010)
Browse files- config.json +72 -0
- generation_config.json +9 -0
- model.safetensors +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +10 -0
config.json
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"model_type": "sn38-nanochrono",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"NanochronoForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_multiplier": 1.2,
|
| 7 |
+
"aux_gate_dim": 12,
|
| 8 |
+
"aux_layers": [
|
| 9 |
+
1,
|
| 10 |
+
3,
|
| 11 |
+
5,
|
| 12 |
+
7,
|
| 13 |
+
9,
|
| 14 |
+
11,
|
| 15 |
+
13,
|
| 16 |
+
15,
|
| 17 |
+
17,
|
| 18 |
+
19,
|
| 19 |
+
21,
|
| 20 |
+
23,
|
| 21 |
+
25,
|
| 22 |
+
27
|
| 23 |
+
],
|
| 24 |
+
"bos_token_id": 32759,
|
| 25 |
+
"eos_token_id": 32763,
|
| 26 |
+
"pad_token_id": 32763,
|
| 27 |
+
"dtype": "bfloat16",
|
| 28 |
+
"final_logit_softcapping": 15.0,
|
| 29 |
+
"hidden_size": 1792,
|
| 30 |
+
"intermediate_size": 7168,
|
| 31 |
+
"layer_types": [
|
| 32 |
+
"sliding_attention",
|
| 33 |
+
"sliding_attention",
|
| 34 |
+
"sliding_attention",
|
| 35 |
+
"full_attention",
|
| 36 |
+
"sliding_attention",
|
| 37 |
+
"sliding_attention",
|
| 38 |
+
"sliding_attention",
|
| 39 |
+
"full_attention",
|
| 40 |
+
"sliding_attention",
|
| 41 |
+
"sliding_attention",
|
| 42 |
+
"sliding_attention",
|
| 43 |
+
"full_attention",
|
| 44 |
+
"sliding_attention",
|
| 45 |
+
"sliding_attention",
|
| 46 |
+
"sliding_attention",
|
| 47 |
+
"full_attention",
|
| 48 |
+
"sliding_attention",
|
| 49 |
+
"sliding_attention",
|
| 50 |
+
"sliding_attention",
|
| 51 |
+
"full_attention",
|
| 52 |
+
"sliding_attention",
|
| 53 |
+
"sliding_attention",
|
| 54 |
+
"sliding_attention",
|
| 55 |
+
"full_attention",
|
| 56 |
+
"sliding_attention",
|
| 57 |
+
"sliding_attention",
|
| 58 |
+
"sliding_attention",
|
| 59 |
+
"full_attention"
|
| 60 |
+
],
|
| 61 |
+
"max_position_embeddings": 2048,
|
| 62 |
+
"mix_gate_dim": 24,
|
| 63 |
+
"num_attention_heads": 14,
|
| 64 |
+
"num_hidden_layers": 28,
|
| 65 |
+
"num_key_value_heads": 14,
|
| 66 |
+
"rope_theta": 100000.0,
|
| 67 |
+
"sliding_window": 512,
|
| 68 |
+
"tap_layer": 14,
|
| 69 |
+
"tie_word_embeddings": false,
|
| 70 |
+
"use_cache": true,
|
| 71 |
+
"vocab_size": 32768
|
| 72 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 32759,
|
| 3 |
+
"eos_token_id": 32763,
|
| 4 |
+
"pad_token_id": 32763,
|
| 5 |
+
"do_sample": false,
|
| 6 |
+
"num_beams": 8,
|
| 7 |
+
"early_stopping": true,
|
| 8 |
+
"length_penalty": 1.0
|
| 9 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d1df02b2ff1c8dc8adc692b7109dc4dfb3260166584c66cf5a0827a668b61b8
|
| 3 |
+
size 4037045748
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|bos|>",
|
| 3 |
+
"eos_token": "<|assistant_end|>",
|
| 4 |
+
"pad_token": "<|assistant_end|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"model_max_length": 2048,
|
| 7 |
+
"add_bos_token": true,
|
| 8 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 9 |
+
"chat_template": "{{ bos_token }}{% set sys = namespace(t='') %}{% for m in messages %}{% if m['role'] == 'system' %}{% set sys.t = m['content'] + '\n\n' %}{% elif m['role'] == 'user' %}<|user_start|>{{ sys.t }}{{ m['content'] }}<|user_end|>{% set sys.t = '' %}{% elif m['role'] == 'assistant' %}<|assistant_start|>{{ m['content'] }}<|assistant_end|>{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant_start|>{% endif %}"
|
| 10 |
+
}
|