INDEX

Explanations

similar

np_max-act · gemini-2.0-flash

The neuron activates on the word “similar” (and its equivalent forms/translations), flagging comparative “similar to…” expressions.

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

 numerator

-0.07

Object

-0.07

τερο

-0.07

rak

-0.06

cco

-0.06

(method

-0.06

 Targets

-0.06

_BUTTON

-0.06

.Temp

-0.06

دث

-0.06

POSITIVE LOGITS

 similar

0.08

istrator

0.07

 ActivityCompat

0.07

 synerg

0.07

 relentlessly

0.06

��

0.06

xbf

0.06

_hold

0.06

히

0.06

 Buffered

0.06

Activations Density 0.059%

similar

The neuron activates on the word “similar” (and its equivalent forms/translations), flagging comparative “similar to…” expressions.

No Comments

No Known Activations