INDEX

Explanations

references to state or status indicators in selection contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

IsContent

-0.97

OGND

-0.97

InjectAttribute

-0.91

 defaultstate

-0.91

 CreateTagHelper

-0.90

 Roskov

-0.87

`,

-0.86

;'>

-0.86

.”)

-0.85

 AssemblyCompany

-0.85

POSITIVE LOGITS

Selected

1.34

 Selected

1.27

selected

1.19

 selected

1.19

SELECTED

1.11

 SELECTED

1.07

setSelected

1.06

getSelected

1.00

 setSelected

0.93

 getSelected

0.92

Activations Density 0.058%

references to state or status indicators in selection contexts

No Comments

No Known Activations