INDEX

Explanations

specific date ranges or timelines in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

tagHelperRunner

-0.60

Autoritní

-0.58

Vidite

-0.56

 ſeveral

-0.54

 بيها

-0.52

 defaultstate

-0.52

SourceChecksum

-0.49

 ſch

-0.48

 ffilm

-0.48

PLIED

-0.48

POSITIVE LOGITS

 until

1.48

until

1.34

 till

1.28

Until

1.23

 Until

1.18

 hasta

1.17

till

1.11

 UNTIL

1.10

 untill

1.04

直到

1.03

Activations Density 0.244%

specific date ranges or timelines in text

No Comments

No Known Activations