INDEX

Explanations

code or programming-related terminology

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

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

fjspx

-0.99

setVerticalGroup

-0.89

MessageTagHelper

-0.81

ंदीखरीदारी

-0.78

تقاوى

-0.75

SharedDtor

-0.73

SourceChecksum

-0.73

ⓧ

-0.72

 onCancelled

-0.72

 Collection

-0.71

POSITIVE LOGITS

Text

0.65

 Text

0.43

ktop

0.43

text

0.41

semantics

0.41

 text

0.41

Wam

0.40

 reservados

0.39

 autora

0.38

Web

0.38

Activations Density 0.046%

code or programming-related terminology

No Comments

No Known Activations