INDEX

Explanations

terms related to shape and form

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 Munro

-0.90

 myſelf

-0.89

CodedInputStream

-0.88

 Monfieur

-0.87

wój

-0.87

soever

-0.87

 acorns

-0.84

 thirds

-0.83

$_"

-0.83

"]));

-0.81

POSITIVE LOGITS

 shape

1.85

 SHAPE

1.80

 shapes

1.73

 Shape

1.71

 Shapes

1.64

shape

1.64

Shape

1.61

SHAPE

1.60

shapes

1.55

Shapes

1.39

Activations Density 0.053%

terms related to shape and form

No Comments

No Known Activations