INDEX

Explanations

coding

np_max-act · gemini-2.0-flash

This neuron detects the JSON‐style markup and keys used for specifying tool actions (e.g. backticks, braces, quotes, and the “action”/“action_input” fields).

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

ific

-0.07

.glide

-0.06

_calls

-0.06

 invading

-0.06

arden

-0.06

릴

-0.06

 Lake

-0.06

 entrusted

-0.06

 Gonzalez

-0.06

BackPressed

-0.06

POSITIVE LOGITS

 Powered

0.06

 Chevron

0.06

چ

0.06

 colourful

0.06

	setTimeout

0.06

 mensen

0.06

 рів

0.06

 rozh

0.06

 musica

0.06

 cheering

0.06

Activations Density 0.011%

coding

This neuron detects the JSON‐style markup and keys used for specifying tool actions (e.g. backticks, braces, quotes, and the “action”/“action_input” fields).

No Comments

No Known Activations