INDEX

Explanations

data types and programming constructs within code snippets

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

Data types (long, double, boolean, String)

np_acts-logits-general · gemini-2.0-flash

variable names and types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 itſelf

-1.15

 myſelf

-1.04

 noastre

-0.98

 Majefty

-0.98

 pleaſure

-0.98

ſelves

-0.97

ſelf

-0.96

 Anſ

-0.95

 perſon

-0.95

 Diſ

-0.94

POSITIVE LOGITS

[]

1.54

[]

1.10

[][]

1.00

et

0.88

aux

0.78

0.77

0.74

0.73

0.72

Activations Density 0.100%

data types and programming constructs within code snippets

Data types (long, double, boolean, String)

variable names and types

No Comments

No Known Activations