INDEX

Explanations

references to groups of people or audiences in various contexts

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

relating to groups of people

np_acts-logits-general · gemini-2.0-flash

people performing actions

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

 ones

-0.50

pro

-0.49

the

-0.48

The

-0.45

—

-0.45

among

-0.44

rov

-0.43

-0.41

po

-0.41

POSITIVE LOGITS

 يتيمه

1.09

ValueStyle

1.00

GenerationType

0.94

 propOrder

0.93

 themſelves

0.87

 الحره

0.86

 ſtate

0.85

évaluateur

0.84

 myſelf

0.83

 ſhould

0.83

Activations Density 0.409%

references to groups of people or audiences in various contexts

relating to groups of people

people performing actions

No Comments

No Known Activations