INDEX

Explanations

phrases indicating uncertainty or conjecture in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tvguidetime

-1.00

脚注の使い方

-0.85

########.

-0.81

Diweddarwch

-0.81

 transfieras

-0.80

yntaxException

-0.75

$_"

-0.75

djangoproject

-0.71

adaptiveStyles

-0.71

Efq

-0.69

POSITIVE LOGITS

 again

0.90

0.78

again

0.76

 Again

0.74

Again

0.74

 then

0.73

 part

0.62

 also

0.59

 particularly

0.58

 certainly

0.58

Activations Density 0.079%

phrases indicating uncertainty or conjecture in discussions

No Comments

No Known Activations