INDEX

Explanations

code

np_max-act · gemini-2.0-flash

The neuron detects Java print‐out code, in particular System.out.println calls.

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

code snippets that involve working with data structures and processing logic.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 notice

-0.06

.factor

-0.06

 Borders

-0.06

(rr

-0.06

جا

-0.06

_CLR

-0.06

 noteworthy

-0.06

NSE

-0.06

CFO

-0.06

 evolutionary

-0.06

POSITIVE LOGITS

 getContext

0.07

.Con

0.06

efs

0.06

.poi

0.06

فة

0.06

fell

0.06

Concept

0.06

 feature

0.06

_Channel

0.06

**

0.06

Activations Density 0.124%

code

The neuron detects Java print‐out code, in particular System.out.println calls.

code snippets that involve working with data structures and processing logic.

No Comments

No Known Activations