INDEX

Explanations

terms related to filling or completion

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+#+

-0.77

 GenerationType

-0.67

AndEndTag

-0.66

 AssemblyProduct

-0.65

}{*}{

-0.63

rawDesc

-0.61

 asupra

-0.60

ніципа

-0.59

 &___

-0.59

存于互联网档案馆

-0.57

POSITIVE LOGITS

filling

1.16

 gaps

1.14

 FILL

1.12

filled

1.08

 fills

1.07

 Fill

1.06

 Gaps

1.05

FILL

1.05

 Filling

1.05

 filling

1.04

Activations Density 0.082%

terms related to filling or completion

No Comments

No Known Activations