INDEX

Explanations

specific references to quantities and measurements in various contexts

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

exact amount or number

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

 <<<<<<<<<<<<<<

-0.54

InitVars

-0.54

etcode

-0.53

 pinulongan

-0.52

verwijspagina

-0.49

isNullOrEmpty

-0.43

MemoryWarning

-0.43

 Indica

-0.42

 saites

-0.42

amai

-0.42

POSITIVE LOGITS

 exactly

1.28

exactly

1.20

 exact

1.16

exact

1.14

Exactly

1.14

 Exactly

1.13

 exactamente

1.11

 exatamente

1.09

 precisely

1.09

 EXACTLY

1.06

Activations Density 0.060%

specific references to quantities and measurements in various contexts

exact amount or number

No Comments

No Known Activations