INDEX

Explanations

references to API calls and related technical terms

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

API calls and 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

 للمعارف

-0.53

rungsseite

-0.53

 dezelve

-0.48

 zelve

-0.47

 spreken

-0.47

 propOrder

-0.45

 goederen

-0.45

 يتيمه

-0.45

 myö

-0.45

 zoude

-0.43

POSITIVE LOGITS

API

2.44

API

1.41

APIs

1.25

WebAPI

1.00

APIS

0.93

 APIs

0.88

APIC

0.81

Api

0.80

SDK

0.71

APN

0.70

Activations Density 0.007%

references to API calls and related technical terms

API calls and user input

No Comments

No Known Activations