INDEX

Explanations

numeric values and their associated labels or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Walpole

-0.54

 AssemblyVersion

-0.52

at

-0.47

 aiut

-0.47

 obiettivi

-0.47

 expansion

-0.46

ところ

-0.45

 esempi

-0.45

 incontri

-0.45

цыя

-0.44

POSITIVE LOGITS

存于互联网档案馆

0.79

ChildScrollView

0.73

ArgsConstructor

0.69

Források

0.67

 BoxDecoration

0.67

 nakalista

0.66

تقاوى

0.65

PyExc

0.63

 للمعارف

0.63

enumii

0.62

Activations Density 0.852%

numeric values and their associated labels or identifiers

No Comments

No Known Activations