INDEX

Explanations

connections and relationships between different concepts or subjects

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

pronouns and names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

RTEE

-0.65

 שוליים

-0.64

RectangleBorder

-0.62

 GenerationType

-0.62

 MessageBoxIcon

-0.61

VersionUID

-0.60

 XCTest

-0.60

AISSEE

-0.59

tableFuture

-0.56

 ddelweddau

-0.56

POSITIVE LOGITS

 viņ

0.38

 they

0.34

 उन्होंने

0.33

 rés

0.33

他们在

0.32

she

0.30

imágenes

0.30

 beliau

0.28

 niej

0.28

 उसने

0.28

Activations Density 0.090%

connections and relationships between different concepts or subjects

pronouns and names

No Comments

No Known Activations