Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -17,32 +17,7 @@ pipeline_tag: text-generation
|
|
| 17 |
qunatized_by: twhoool02
|
| 18 |
---
|
| 19 |
|
| 20 |
-
# Model Card for
|
| 21 |
-
(model): LlamaModel(
|
| 22 |
-
(embed_tokens): Embedding(32000, 4096)
|
| 23 |
-
(layers): ModuleList(
|
| 24 |
-
(0-31): 32 x LlamaDecoderLayer(
|
| 25 |
-
(self_attn): LlamaSdpaAttention(
|
| 26 |
-
(rotary_emb): LlamaRotaryEmbedding()
|
| 27 |
-
(k_proj): QuantLinear()
|
| 28 |
-
(o_proj): QuantLinear()
|
| 29 |
-
(q_proj): QuantLinear()
|
| 30 |
-
(v_proj): QuantLinear()
|
| 31 |
-
)
|
| 32 |
-
(mlp): LlamaMLP(
|
| 33 |
-
(act_fn): SiLU()
|
| 34 |
-
(down_proj): QuantLinear()
|
| 35 |
-
(gate_proj): QuantLinear()
|
| 36 |
-
(up_proj): QuantLinear()
|
| 37 |
-
)
|
| 38 |
-
(input_layernorm): LlamaRMSNorm()
|
| 39 |
-
(post_attention_layernorm): LlamaRMSNorm()
|
| 40 |
-
)
|
| 41 |
-
)
|
| 42 |
-
(norm): LlamaRMSNorm()
|
| 43 |
-
)
|
| 44 |
-
(lm_head): Linear(in_features=4096, out_features=32000, bias=False)
|
| 45 |
-
)
|
| 46 |
|
| 47 |
## Model Details
|
| 48 |
|
|
@@ -51,32 +26,7 @@ This model is a GPTQ quantized version of the meta-llama/Llama-2-7b-hf model.
|
|
| 51 |
- **Developed by:** Ted Whooley
|
| 52 |
- **Library:** Transformers, GPTQ
|
| 53 |
- **Model type:** llama
|
| 54 |
-
- **Model name:**
|
| 55 |
-
(model): LlamaModel(
|
| 56 |
-
(embed_tokens): Embedding(32000, 4096)
|
| 57 |
-
(layers): ModuleList(
|
| 58 |
-
(0-31): 32 x LlamaDecoderLayer(
|
| 59 |
-
(self_attn): LlamaSdpaAttention(
|
| 60 |
-
(rotary_emb): LlamaRotaryEmbedding()
|
| 61 |
-
(k_proj): QuantLinear()
|
| 62 |
-
(o_proj): QuantLinear()
|
| 63 |
-
(q_proj): QuantLinear()
|
| 64 |
-
(v_proj): QuantLinear()
|
| 65 |
-
)
|
| 66 |
-
(mlp): LlamaMLP(
|
| 67 |
-
(act_fn): SiLU()
|
| 68 |
-
(down_proj): QuantLinear()
|
| 69 |
-
(gate_proj): QuantLinear()
|
| 70 |
-
(up_proj): QuantLinear()
|
| 71 |
-
)
|
| 72 |
-
(input_layernorm): LlamaRMSNorm()
|
| 73 |
-
(post_attention_layernorm): LlamaRMSNorm()
|
| 74 |
-
)
|
| 75 |
-
)
|
| 76 |
-
(norm): LlamaRMSNorm()
|
| 77 |
-
)
|
| 78 |
-
(lm_head): Linear(in_features=4096, out_features=32000, bias=False)
|
| 79 |
-
)
|
| 80 |
- **Pipeline tag:** text-generation
|
| 81 |
- **Qunatized by:** twhoool02
|
| 82 |
- **Language(s) (NLP):** en
|
|
|
|
| 17 |
qunatized_by: twhoool02
|
| 18 |
---
|
| 19 |
|
| 20 |
+
# Model Card for twhoool02/Llama-2-7b-hf-AutoGPTQ
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 21 |
|
| 22 |
## Model Details
|
| 23 |
|
|
|
|
| 26 |
- **Developed by:** Ted Whooley
|
| 27 |
- **Library:** Transformers, GPTQ
|
| 28 |
- **Model type:** llama
|
| 29 |
+
- **Model name:** Llama-2-7b-hf-AutoGPTQ
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
- **Pipeline tag:** text-generation
|
| 31 |
- **Qunatized by:** twhoool02
|
| 32 |
- **Language(s) (NLP):** en
|