INDEX

Explanations

numerical values and mathematical expressions, particularly related to linear equations and calibration models

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

.."

-0.62

-"

-0.55

Ã

-0.55

$=$

-0.54

Mar

-0.53

$\

-0.53

$>$

-0.52

$*$

-0.51

**

-0.51

$($

-0.51

POSITIVE LOGITS

 uſed

0.99

 Monfieur

0.97

 purpoſe

0.97

 becauſe

0.96

 ſtand

0.93

 ſtill

0.91

 themſelves

0.91

 ſta

0.90

 reaſon

0.90

 ainfi

0.89

Activations Density 1.364%

numerical values and mathematical expressions, particularly related to linear equations and calibration models

No Comments

No Known Activations