INDEX

Explanations

words related to emotions or feelings

oai_token-act-pair · gemini-2.0-flash

abstract or complex concepts

np_token-act-pair-logits · gpt-4o-mini

internal thoughts/feelings

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ywna

-0.49

arası

-0.48

 tanken

-0.45

icide

-0.44

ISupport

-0.42

olate

-0.42

 Linie

-0.42

Попис

-0.42

ibatis

-0.42

 heaped

-0.41

POSITIVE LOGITS

 Roskov

0.77

 незавершена

0.72

+#+#

0.70

 kaynağından

0.67

featureID

0.67

oredCriteria

0.66

Personendaten

0.66

KURZBESCHREIBUNG

0.65

LikeLiked

0.65

rungsseite

0.65

Activations Density 1.492%

words related to emotions or feelings

abstract or complex concepts

internal thoughts/feelings

No Comments

No Known Activations