INDEX

Explanations

color-related terms

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-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

red

-1.02

 black

-0.99

 white

-0.95

 blue

-0.80

white

-0.70

 green

-0.70

 brown

-0.68

 grey

-0.68

 yellow

-0.68

black

-0.68

POSITIVE LOGITS

 typelib

0.81

:✨

0.78

@"/

0.75

 &___

0.65

oa̍t

0.63

TargetException

0.61

0.60

stdc

0.59

 חיצוניים

0.59

onade

0.58

Activations Density 0.195%

color-related terms

No Comments

No Known Activations