INDEX

Explanations

,

np_max-act · gemini-2.0-flash

specific names or designations in a text.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

The neuron fires strongly on long, multi‐segment compound words (e.g. repeated subword fragments in German compounds).

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

 them

-0.07

 расп

-0.07

ेकर

-0.06

taj

-0.06

략

-0.06

 인정

-0.06

emann

-0.06

 návště

-0.06

'-

-0.06

POSITIVE LOGITS

.Receive

0.08

 времени

0.07

[string

0.07

Response

0.07

blocking

0.07

ipheral

0.06

_minute

0.06

[length

0.06

<Address

0.06

_FULLSCREEN

0.06

Activations Density 0.009%

,

specific names or designations in a text.

The neuron fires strongly on long, multi‐segment compound words (e.g. repeated subword fragments in German compounds).

No Comments

No Known Activations

,

specific names or designations in a text.

The neuron fires strongly on long, multi‐segment compound words (e.g. repeated subword fragments in German compounds).

No Comments

No Known Activations