INDEX

Explanations

programming constructs involving logical conditions and array manipulations

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

Code, symbols, or delimiters

np_acts-logits-general · gemini-2.0-flash

testDateopensite

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Efq

-1.24

 Monfieur

-1.12

 itſelf

-1.09

 iſt

-1.09

 myſelf

-1.06

 Jefus

-1.02

 Houſe

-1.01

 auffi

-1.00

 ſind

-0.99

 houſe

-0.97

POSITIVE LOGITS

0.70

le

0.67

non

0.66

↵

0.65

<eos>

0.64

↵↵

0.64

no

0.61

No

0.60

Activations Density 0.353%

programming constructs involving logical conditions and array manipulations

Code, symbols, or delimiters

testDateopensite

No Comments

No Known Activations