INDEX

Explanations

code-related constructs and operations in JSON data processing

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

get("key") or user input

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

 Perci

-0.34

 perang

-0.33

ispo

-0.33

 spoil

-0.33

 maceta

-0.32

freiheit

-0.31

 Pernambuco

-0.31

latego

-0.31

writeFieldEnd

-0.31

uação

-0.30

POSITIVE LOGITS

 surla

0.72

getJSONObject

0.56

GOTREF

0.56

 nakalista

0.53

hoeddwyd

0.53

mybatisplus

0.53

 defaultstate

0.52

 PhpStorm

0.51

IsContent

0.50

intios

0.49

Activations Density 0.009%

code-related constructs and operations in JSON data processing

get("key") or user input

No Comments

No Known Activations