INDEX

Explanations

terms related to software, programming, and technical specifications

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

Follows the beginning of a word

np_acts-logits-general · gemini-2.0-flash

named or called

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

chartInstance

-0.48

complexContent

-0.44

antMatchers

-0.42

RepeatedField

-0.41

 ayuntamiento

-0.41

GHIJKLM

-0.39

 AssemblyCulture

-0.39

tslib

-0.37

Demografia

-0.37

gifter

-0.36

POSITIVE LOGITS

 named

1.30

 called

1.27

名为

1.16

 bernama

1.15

 Called

1.11

named

1.11

Called

1.09

名叫

1.09

 titled

1.05

called

1.04

Activations Density 0.980%

terms related to software, programming, and technical specifications

Follows the beginning of a word

named or called

No Comments

No Known Activations