INDEX

Explanations

references to a wide range of options or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 על

-0.53

 tempi

-0.51

podar

-0.49

 hmot

-0.47

igl

-0.47

:+:

-0.46

tréal

-0.46

 Berech

-0.46

ulio

-0.46

Nan

-0.46

POSITIVE LOGITS

 variety

2.53

variety

2.31

 Variety

2.18

Variety

2.08

 varieties

1.93

 Varieties

1.82

 VARI

1.82

VARI

1.74

 variedad

1.74

 variété

1.70

Activations Density 0.149%

references to a wide range of options or categories

No Comments

No Known Activations