INDEX

Explanations

code or programming constructs related to defining and managing types or data structures

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

Code and programming-related terms

np_acts-logits-general · gemini-2.0-flash

programming keywords and Python

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

ExecuteReader

-0.50

saraba

-0.50

EnableWeb

-0.48

 StringTokenizer

-0.48

MeasureSpec

-0.47

piration

-0.46

ifstream

-0.46

 Colorado

-0.45

EndGlobal

-0.45



-0.45

POSITIVE LOGITS

Py

1.14

 PyObject

1.12

Py

1.09

pyx

1.03

PyObject

0.93

PY

0.81

 Python

0.81

Python

0.80

NPY

0.79

 python

0.77

Activations Density 0.078%

code or programming constructs related to defining and managing types or data structures

Code and programming-related terms

programming keywords and Python

No Comments

No Known Activations