INDEX

Explanations

relationships and interactions among characters

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

Pronouns referring to people

np_acts-logits-general · gemini-2.0-flash

subject followed by verb

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

got

-0.91

 gonna

-0.86

 going

-0.81

get

-0.81

 saying

-0.80

 messed

-0.75

 gets

-0.74

 really

-0.74

 REALLY

-0.73

go

-0.72

POSITIVE LOGITS

 itſelf

0.81

 myſelf

0.81

 swiftly

0.75

]--;

0.75

 propOrder

0.74

DoubleQuotes

0.73

]),

0.73

 foon

0.73

 himſelf

0.72

 poffe

0.71

Activations Density 1.425%

relationships and interactions among characters

Pronouns referring to people

subject followed by verb

No Comments

No Known Activations