INDEX

Explanations

references to the word "record" and its variations, indicating topics related to documentation and data retention

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

"record" and its surrounding context

np_acts-logits-general · gemini-2.0-flash

record keeping and breaking

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 "..\..\

-0.80

Datuak

-0.77

RepeatedField

-0.75

pherals

-0.75

DispatchToProps

-0.74

Искәрмәләр

-0.74

 CreateTagHelper

-0.72

 GetEnumerator

-0.71

 "..\..\..\

-0.70

BeginContext

-0.69

POSITIVE LOGITS

keeping

0.66

setting

0.66

breaking

0.65

atorios

0.63

book

0.62

records

0.61

Records

0.60

nungs

0.60

record

0.58

 RECORD

0.58

Activations Density 0.125%

references to the word "record" and its variations, indicating topics related to documentation and data retention

"record" and its surrounding context

record keeping and breaking

No Comments

No Known Activations