INDEX

Explanations

spatial and temporal relationship indicators in the text

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

before, after, around, below, above

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

EndInit

-0.57

+#+

-0.56

IBOutlet

-0.56

SequentialGroup

-0.54

exitRule

-0.50

BeginInit

-0.50

 nahilalakip

-0.50

揄

-0.50

 oprot

-0.49

fetchall

-0.49

POSITIVE LOGITS

 виправивши

0.45

Климат

0.34

shell

0.34

LookAnd

0.33

BagConstraints

0.33

 barras

0.32

ędzy

0.32

SpringRunner

0.32

spaces

0.32

 minas

0.32

Activations Density 0.037%

spatial and temporal relationship indicators in the text

before, after, around, below, above

No Comments

No Known Activations