INDEX

Explanations

the

np_max-act · gemini-2.0-flash

The neuron detects phrases describing material or streams being sent back or recirculated to an earlier process stage (e.g. “back to the feed,” “recirculated back to the flotation step”).

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

Negative Logits

čast

-0.06

 cryptocurrencies

-0.06

 approached

-0.06

 Biol

-0.06

 Bowen

-0.06

ляти

-0.06

.Elapsed

-0.06

MG

-0.06

.En

-0.06

pell

-0.06

POSITIVE LOGITS

Sharp

0.06

Ped

0.06

-target

0.06

 işlemi

0.06

.uint

0.06

 removeObject

0.06

SEA

0.06

nos

0.06

 gửi

0.06

SPR

0.06

Activations Density 0.012%

the

The neuron detects phrases describing material or streams being sent back or recirculated to an earlier process stage (e.g. “back to the feed,” “recirculated back to the flotation step”).

No Known Activations

the

The neuron detects phrases describing material or streams being sent back or recirculated to an earlier process stage (e.g. “back to the feed,” “recirculated back to the flotation step”).

No Known Activations