INDEX

Explanations

phrases indicating sudden changes or interruptions

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

 للمعارف

-0.86

 متعلقه

-0.82

__*/

-0.81

 FetchType

-0.80

aarrggbb

-0.80

 invokingState

-0.77

Portale

-0.76

 ostavi

-0.75

 defStyle

-0.75

tagHelperRunner

-0.73

POSITIVE LOGITS

 suddenly

1.52

 sudden

1.44

 Suddenly

1.25

Suddenly

1.19

suddenly

1.08

all

1.06

Sudden

1.06

 Sudden

1.05

sud

0.96

sud

0.92

Activations Density 0.145%

phrases indicating sudden changes or interruptions

No Comments

No Known Activations