INDEX

Explanations

verbs and phrases that indicate attempts or plans to assess situations

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

makeConstraints

-0.74

 BoxDecoration

-0.73

WithIOException

-0.71

 يتيمه

-0.66

 propOrder

-0.66

 betweenstory

-0.65

StoryboardSegue

-0.64

AxisAlignment

-0.62

 InputDecoration

-0.61

Achtung

-0.59

POSITIVE LOGITS

see

1.87

See

1.59

See

1.50

see

1.45

 sees

1.29

 voir

1.28

SEE

1.24

看看

1.17

 seeing

1.17

xem

1.15

Activations Density 0.234%

verbs and phrases that indicate attempts or plans to assess situations

No Comments

No Known Activations