INDEX

Explanations

terms and phrases related to visual characteristics and analysis

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

Something being seen/observed

np_acts-logits-general · gemini-2.0-flash

describing how things look

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

abestanden

-0.84

 propOrder

-0.79

AddTagHelper

-0.70

 defaultstate

-0.65

ValueGeneration

-0.62

\{\\

-0.61

GenerationType

-0.61

providedIn

-0.58

DialogContent

-0.57

ThroughAttribute

-0.57

POSITIVE LOGITS

 viewed

0.93

 when

0.87

 Viewed

0.81

viewed

0.72

Viewed

0.70

 vistos

0.69

when

0.65

 viewing

0.65

 zoomed

0.64

 cuando

0.64

Activations Density 0.425%

terms and phrases related to visual characteristics and analysis

Something being seen/observed

describing how things look

No Comments

No Known Activations