INDEX

Explanations

attends to action-related tokens associated with instructional contexts from enforceable tokens indicating conditions or restrictions

oai_attention-head · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-att-16k

Configuration

google/gemma-scope-2b-pt-att/layer_10/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.attn.hook_z

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.06

1:0.07

2:0.04

3:0.06

4:0.05

5:0.03

6:0.56

7:0.09

Negative Logits

 ویکی‌پدیا

-0.43

BufferException

-0.33

Vidite

-0.31

थि

-0.31

ageo

-0.31

MemoryWarning

-0.31

dicionado

-0.31

KommentareTeilen

-0.30

prüche

-0.30

</em>

-0.30

POSITIVE LOGITS

}))

0.39

DockStyle

0.38

 transfieras

0.36

 confirmer

0.36

.)}

0.36

ρους

0.36

غيرة

0.35

}])

0.34

 Reihe

0.34

ianum

0.34

Activations Density 0.096%

attends to action-related tokens associated with instructional contexts from enforceable tokens indicating conditions or restrictions

No Comments

No Known Activations