INDEX

Explanations

references to looking back at past experiences or memories

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

looking back

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

宜

-0.33

 تضيفلها

-0.33

Spoljašnje

-0.30

 المو

-0.30

翻攝

-0.30

Sap

-0.30

 Magist

-0.29

romptu

-0.28

rem

-0.28

 aang

-0.28

POSITIVE LOGITS

 retrospective

0.74

 retrospec

0.70

振り返

0.69

 recalling

0.68

回顾

0.67

 recalls

0.67

SequentialGroup

0.66

回想

0.66

 recollections

0.66

 Retrospective

0.66

Activations Density 0.003%

references to looking back at past experiences or memories

looking back

No Comments

No Known Activations