INDEX

Explanations

content related to emotional expressions and physical reactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

AddTagHelper

-0.61

Datuak

-0.57

ignite

-0.53

 تانيه

-0.52

UseVisualStyle

-0.48

atorship

-0.47

 doInBackground

-0.45

 Heel

-0.45

 Heels

-0.44

zęd

-0.44

POSITIVE LOGITS

 face

2.65

 faces

2.39

 Face

2.33

face

2.30

Face

2.23

 FACE

2.10

 facial

2.07

 Faces

2.02

faces

1.97

FACE

1.89

Activations Density 0.187%

content related to emotional expressions and physical reactions

No Comments

No Known Activations