INDEX

Explanations

references to different types and characteristics of spectra

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

 Firm

-0.53

big

-0.52

Thi

-0.51

Fir

-0.51

JB

-0.51

 Home

-0.51

home

-0.50

JB

-0.50

 firm

-0.48

“

-0.47

POSITIVE LOGITS

Spectrum

1.03

 spectrum

1.03

 Spectrum

1.02

spectrum

0.96

 SPECT

0.94

 Spectra

0.94

 spectra

0.90

 spectral

0.79

spectra

0.78

 Spectre

0.77

Activations Density 0.018%

references to different types and characteristics of spectra

No Comments

No Known Activations