INDEX

Explanations

references to political entities and events

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

 used

-0.30

ait

-0.29

 flew

-0.28

des

-0.28

 coll

-0.28

خته

-0.28

-0.27

bo

-0.27

POSITIVE LOGITS

 pleaſure

0.86

 desmotivaciones

0.81

 queſta

0.79

phosa

0.79

 otomatig

0.78

reportWebVitals

0.78

 kasarigan

0.78

 Numerade

0.77

 faſt

0.76

 Normdatei

0.73

Activations Density 0.009%

references to political entities and events

No Comments

No Known Activations