INDEX

Explanations

terms related to models and their attributes or evaluations

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

model parameters and return values

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

The neuron detects document-structure labels and section/headline-like tokens (e.g., short capitalized or header words such as "Model", "Parameter", "Control", "return", "market risk premium").

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 dotyczące

-0.28

ยัง

-0.27

连忙

-0.26

 sportowe

-0.26

 experto

-0.26

 otherwise

-0.25

 equally

-0.25

เหร

-0.24

 koń

-0.23

 exciting

-0.22

POSITIVE LOGITS

Tembelea

0.79

AndEndTag

0.77

 Infórmanos

0.74

出版年

0.71

 EconPapers

0.71

 समीक्षक

0.71

 '\\;'

0.68

strophy

0.68

 OFDb

0.67

 للمعارف

0.66

Activations Density 0.000%

terms related to models and their attributes or evaluations

model parameters and return values

The neuron detects document-structure labels and section/headline-like tokens (e.g., short capitalized or header words such as "Model", "Parameter", "Control", "return", "market risk premium").

No Comments

No Known Activations

terms related to models and their attributes or evaluations

model parameters and return values

The neuron detects document-structure labels and section/headline-like tokens (e.g., short capitalized or header words such as "Model", "Parameter", "Control", "return", "market risk premium").

No Comments

No Known Activations