INDEX

Explanations

references to singular and significant aspects or entities

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

Follows "not a" or "not"

np_acts-logits-general · gemini-2.0-flash

a common noun

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

 autorytatywna

-0.77

Various

-0.76

almost

-0.73

 Various

-0.72

various

-0.69

 various

-0.68

 almost

-0.67

 VARIOUS

-0.65

 verschiedenen

-0.64

 einiges

-0.63

POSITIVE LOGITS

 anymore

1.42

 anything

1.35

any

1.34

nor

1.28

 quelconque

1.27

anything

1.13

 nici

1.12

 żad

1.06

 slightest

1.06

 necessarily

1.04

Activations Density 0.307%

references to singular and significant aspects or entities

Follows "not a" or "not"

a common noun

No Comments

No Known Activations