Upload checkpoint step-00000500
Browse files- config.json +91 -0
- state.json +1 -0
config.json
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architecture": "interleaved",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"TextLMModel"
|
| 5 |
+
],
|
| 6 |
+
"assistant_offset": 8192,
|
| 7 |
+
"attention_bias": false,
|
| 8 |
+
"attention_dropout": 0.0,
|
| 9 |
+
"attn_output_gate": true,
|
| 10 |
+
"blank_loss_weight": 0.1,
|
| 11 |
+
"blank_token_id": 248077,
|
| 12 |
+
"bos_token_id": null,
|
| 13 |
+
"cross_attention_causal": true,
|
| 14 |
+
"decoder_num_hidden_layers": null,
|
| 15 |
+
"dtype": "float32",
|
| 16 |
+
"encoder_num_hidden_layers": null,
|
| 17 |
+
"eos_token_id": 248044,
|
| 18 |
+
"full_attention_interval": 4,
|
| 19 |
+
"head_dim": 256,
|
| 20 |
+
"hidden_act": "silu",
|
| 21 |
+
"hidden_size": 1024,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"intermediate_size": 3584,
|
| 24 |
+
"layer_types": [
|
| 25 |
+
"linear_attention",
|
| 26 |
+
"linear_attention",
|
| 27 |
+
"linear_attention",
|
| 28 |
+
"full_attention",
|
| 29 |
+
"linear_attention",
|
| 30 |
+
"linear_attention",
|
| 31 |
+
"linear_attention",
|
| 32 |
+
"full_attention",
|
| 33 |
+
"linear_attention",
|
| 34 |
+
"linear_attention",
|
| 35 |
+
"linear_attention",
|
| 36 |
+
"full_attention",
|
| 37 |
+
"linear_attention",
|
| 38 |
+
"linear_attention",
|
| 39 |
+
"linear_attention",
|
| 40 |
+
"full_attention",
|
| 41 |
+
"linear_attention",
|
| 42 |
+
"linear_attention",
|
| 43 |
+
"linear_attention",
|
| 44 |
+
"full_attention",
|
| 45 |
+
"linear_attention",
|
| 46 |
+
"linear_attention",
|
| 47 |
+
"linear_attention",
|
| 48 |
+
"full_attention"
|
| 49 |
+
],
|
| 50 |
+
"linear_conv_kernel_dim": 4,
|
| 51 |
+
"linear_key_head_dim": 128,
|
| 52 |
+
"linear_num_key_heads": 16,
|
| 53 |
+
"linear_num_value_heads": 16,
|
| 54 |
+
"linear_value_head_dim": 128,
|
| 55 |
+
"mamba_ssm_dtype": "float32",
|
| 56 |
+
"mask_prob": 0.0,
|
| 57 |
+
"mask_token_id": 248081,
|
| 58 |
+
"max_position_embeddings": 4194304,
|
| 59 |
+
"mlp_only_layers": [],
|
| 60 |
+
"model_type": "text_lm",
|
| 61 |
+
"mtp_num_hidden_layers": 1,
|
| 62 |
+
"mtp_use_dedicated_embeddings": false,
|
| 63 |
+
"num_attention_heads": 8,
|
| 64 |
+
"num_hidden_layers": 24,
|
| 65 |
+
"num_key_value_heads": 2,
|
| 66 |
+
"onset_loss_weight": 0.0,
|
| 67 |
+
"onset_positive_weight": 20.0,
|
| 68 |
+
"pad_token_id": null,
|
| 69 |
+
"partial_rotary_factor": 0.25,
|
| 70 |
+
"random_prob": 0.0,
|
| 71 |
+
"random_token_vocab_size": null,
|
| 72 |
+
"rms_norm_eps": 1e-06,
|
| 73 |
+
"rope_parameters": {
|
| 74 |
+
"mrope_interleaved": true,
|
| 75 |
+
"mrope_section": [
|
| 76 |
+
11,
|
| 77 |
+
11,
|
| 78 |
+
10
|
| 79 |
+
],
|
| 80 |
+
"partial_rotary_factor": 0.25,
|
| 81 |
+
"rope_theta": 10000000,
|
| 82 |
+
"rope_type": "default"
|
| 83 |
+
},
|
| 84 |
+
"shift_smear_eps": 0.1,
|
| 85 |
+
"shift_tolerance": 2,
|
| 86 |
+
"tie_word_embeddings": true,
|
| 87 |
+
"transformers_version": "5.16.1",
|
| 88 |
+
"use_cache": false,
|
| 89 |
+
"user_loss_weight": 1.0,
|
| 90 |
+
"vocab_size": 248320
|
| 91 |
+
}
|
state.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"global_step": 500}
|