INDEX

Explanations

relative clauses

np_max-act · gemini-2.0-flash

This neuron detects tokens that introduce or occur in relative‐clause constructions (e.g. “which,” “is,” “are,” “have”) that link descriptions in technical prose.

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Eval

-0.07

 Else

-0.07

Category

-0.06

 otros

-0.06

 epoxy

-0.06

 Preis

-0.06

krát

-0.06

oreal

-0.06

ken

-0.06

_training

-0.06

POSITIVE LOGITS

.reloadData

0.07

scriptions

0.07

 Modify

0.07

0.06

ileges

0.06

 getTime

0.06

.SetInt

0.06

भग

0.06

 peru

0.06

/ph

0.06

Activations Density 0.071%

relative clauses

This neuron detects tokens that introduce or occur in relative‐clause constructions (e.g. “which,” “is,” “are,” “have”) that link descriptions in technical prose.

No Comments

No Known Activations