INDEX

Explanations

potion

np_max-act · gemini-2.0-flash

The neuron fires on code‐style identifiers and other technical or jargon terms (e.g. filenames and library names like “zap.c” or “Snap.Types,” plus domain-specific words such as “VR,” “hardware,” etc.).

oai_token-act-pair · o4-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

名稱

-0.07

ันทร

-0.07

 crash

-0.06

>();↵↵

-0.06

IRS

-0.06

cü

-0.06

 tuple

-0.06

 flere

-0.06

quant

-0.06

 expenditure

-0.06

POSITIVE LOGITS

(KEY

0.06

┴

0.06

:X

0.06

.onNext

0.06

?('

0.06

 dolay

0.06

 مشک

0.06

|"

0.06

_failure

0.06

 BufferedWriter

0.06

Activations Density 0.184%

potion

The neuron fires on code‐style identifiers and other technical or jargon terms (e.g. filenames and library names like “zap.c” or “Snap.Types,” plus domain-specific words such as “VR,” “hardware,” etc.).

No Comments

No Known Activations

potion

The neuron fires on code‐style identifiers and other technical or jargon terms (e.g. filenames and library names like “zap.c” or “Snap.Types,” plus domain-specific words such as “VR,” “hardware,” etc.).

No Comments

No Known Activations