INDEX

Explanations

programming-related keywords and language constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

__(/*!

-0.63



-0.62

 outcrops

-0.60

esgue

-0.59

 martyrdom

-0.58

 outcrop

-0.57

 Polity

-0.57

 solicitor

-0.56

 mitotic

-0.56

melen

-0.55

POSITIVE LOGITS

下载附件

0.67

出版年

0.66

reportWebVitals

0.65

Erreferentziak

0.64

MockMvc

0.63

DockStyle

0.62

 estekak

0.60

WriteTagHelper

0.60

mybatisplus

0.60

BufferException

0.57

Activations Density 0.150%

programming-related keywords and language constructs

No Comments

No Known Activations