INDEX

Explanations

references to nighttime or nocturnal themes

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 AssemblyCompany

-0.77

ніципалі

-0.71

 transfieras

-0.68

rungsseite

-0.66

 propOrder

-0.63

 autorytatywna

-0.62

migrationBuilder

-0.62

OGND

-0.60

MessageTagHelper

-0.59

ukone

-0.58

POSITIVE LOGITS

 night

1.16

 nighttime

1.09

 nocturn

1.02

Night

1.00

 Night

0.99

 nocturnal

0.99

 NIGHT

0.96

 nocturno

0.91

night

0.91

 nights

0.87

Activations Density 0.026%

references to nighttime or nocturnal themes

No Comments

No Known Activations