INDEX

Explanations

emotional expressions and reactions in narratives

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

descriptive words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ագրություններ

-0.63

 دیکھیے

-0.54

 autorytatywna

-0.48

orsese

-0.48

itinéraire

-0.47

 Haye

-0.47

FormTagHelper

-0.46

 ligiloj

-0.46

 يتيمه

-0.45

InitVars

-0.45

POSITIVE LOGITS

arraycopy

0.42

ailleurs

0.41

appartamento

0.40

ExecuteReader

0.38

 resourceCulture

0.37

ⓧ

0.37

0.35

CppMethod

0.35

sounding

0.35

läng

0.34

Activations Density 0.067%

emotional expressions and reactions in narratives

descriptive words

No Comments

No Known Activations