INDEX

Explanations

references to questioning or inquiry, particularly using the word "Who."

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

ある

-0.55

 former

-0.54

 second

-0.54

chera

-0.52

rakech

-0.52

CRP

-0.51

lans

-0.51

lej

-0.51

 circo

-0.51

 particular

-0.49

POSITIVE LOGITS

Who

2.07

Who

2.06

who

1.85

who

1.84

WHO

1.72

 hvem

1.70

WHO

1.68

 quién

1.54

quién

1.49

Siapa

1.49

Activations Density 0.078%

references to questioning or inquiry, particularly using the word "Who."

No Comments

No Known Activations