INDEX

Explanations

references to the programming language "Python" and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

ofman

-0.68

 alignItems

-0.62

 تضيفلها

-0.60

szönöm

-0.60

 Humanity

-0.59

новниш

-0.58

humanité

-0.58

interopRequire

-0.57

*.

-0.57

CURLOPT

-0.56

POSITIVE LOGITS

Py

1.14

py

1.01

Py

1.01

pym

0.79

py

0.78

PY

0.76

 PyObject

0.73

PY

0.72

 Python

0.68

 pyrite

0.62

Activations Density 0.092%

references to the programming language "Python" and its variants

No Comments

No Known Activations