INDEX

Explanations

references to memory and remembrance

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iblichen

-0.52

Geographie

-0.47

RectangleBorder

-0.45

prong

-0.44

endfor

-0.44

薮

-0.44

 enclose

-0.42

blitz

-0.42

poles

-0.42

epres

-0.42

POSITIVE LOGITS

 memory

0.60

 memories

0.50

 unforgettable

0.45

pushFollow

0.44

memory

0.44

 memorable

0.43

 MEMORY

0.42

stateMutability

0.40

 jamais

0.39

 remembered

0.39

Activations Density 0.067%

references to memory and remembrance

No Comments

No Known Activations