INDEX

Explanations

elements related to data processing and web interactions

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

Tokens after "the"

np_acts-logits-general · gemini-2.0-flash

programming actions and pages

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 stuff

-0.60

égard

-0.55

 smtplib

-0.48

engertian

-0.48

IntoConstraints

-0.47

stuff

-0.47

 coisas

-0.45

Stuff

-0.45

 things

-0.44

 Waray

-0.43

POSITIVE LOGITS

 database

0.75

RegressionTest

0.73

Попис

0.67

 parent

0.65

큼

0.64

MigrationBuilder

0.63

buttonBar

0.63

 EconPapers

0.62

Elsewhere

0.62

 Elsewhere

0.61

Activations Density 0.702%

elements related to data processing and web interactions

Tokens after "the"

programming actions and pages

No Comments

No Known Activations