INDEX

Explanations

references to hierarchical or organizational structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

paravant

-0.57

niająca

-0.55

#+#

-0.54

elfde

-0.54

Rhestr

-0.51

 VIAF

-0.49

有名な

-0.48

édrale

-0.47

SourceChecksum

-0.47

 known

-0.45

POSITIVE LOGITS

new

1.34

new

1.12

 nueva

1.10

 nieuwe

1.08

 nuevos

1.07

新的

1.07

新

1.06

 nuevas

1.04

 baru

1.03

ใหม่

1.02

Activations Density 0.410%

references to hierarchical or organizational structures

No Comments

No Known Activations