INDEX

Explanations

emotional states and descriptors related to intense feelings or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 Wiktionnaire

-0.63

 referenties

-0.58

 FormGroup

-0.58

gridx

-0.57

 Spro

-0.56

 CreateTagHelper

-0.55

 Référence

-0.53

opold

-0.53

 ModelExpression

-0.53

wikk

-0.52

POSITIVE LOGITS

 disturbing

1.26

 captivating

1.25

 compelling

1.19

 frightening

1.19

 gratifying

1.19

 depressing

1.19

 distressing

1.19

 enticing

1.16

 frustrating

1.16

 shocking

1.15

Activations Density 0.317%

emotional states and descriptors related to intense feelings or experiences

No Comments

No Known Activations