INDEX

Explanations

locations and addresses within a specific context

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

house numbers and street names

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tagHelperRunner

-0.77

InitVars

-0.51

ագրություններ

-0.50

 createSlice

-0.47

RTRS

-0.41

IsMutable

-0.41

oredCriteria

-0.41

mergeFrom

-0.40

aarrggbb

-0.40

WithMany

-0.39

POSITIVE LOGITS

Address

0.66

 Address

0.60

address

0.59

 address

0.58

 residencial

0.55

 apartment

0.53

 apartments

0.53

 Street

0.52

 vivienda

0.52

Adresse

0.52

Activations Density 0.008%

locations and addresses within a specific context

house numbers and street names

No Comments

No Known Activations