INDEX

Explanations

phrases indicating emotional states and social interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Personensuche

-0.50

aarrggbb

-0.48

webElementXpaths

-0.45

flashdata

-0.45

yntaxException

-0.45

DotNetBar

-0.44

lexikon

-0.43

Exception

-0.42

queryInterface

-0.42

jda

-0.42

POSITIVE LOGITS

 actually

2.78

actually

2.63

 fact

2.54

 Actually

2.45

 indeed

2.44

Actually

2.42

indeed

2.21

 infatti

2.19

Indeed

2.15

 Indeed

2.10

Activations Density 0.502%

phrases indicating emotional states and social interactions

No Comments

No Known Activations