INDEX

Explanations

references to specific acronyms or terms related to electrical or electronic components

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

elisabeth, elite, elicits

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

principalColumn

-0.56

 picioare

-0.54

isler

-0.49

 aéreo

-0.47

 correto

-0.46

emballage

-0.46

 ibunya

-0.46

 podjet

-0.45

 păr

-0.45

 zprávy

-0.45

POSITIVE LOGITS

EL

1.81

EL

1.52

ELM

1.07

ELS

0.89

ELI

0.83

ELT

0.79

AEL

0.76

Eli

0.75

UL

0.75

Eli

0.74

Activations Density 0.008%

references to specific acronyms or terms related to electrical or electronic components

elisabeth, elite, elicits

No Comments

No Known Activations