INDEX

Explanations

phrases that indicate past events or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

To

-0.51

To

-0.49

Pre

-0.49

目を

-0.48

and

-0.48

 spend

-0.47

Nor

-0.47

Pre

-0.47

хра

-0.46

pre

-0.46

POSITIVE LOGITS

 previous

1.96

 earlier

1.96

earlier

1.86

 anteriormente

1.69

previous

1.66

 Previously

1.62

 Earlier

1.61

 previously

1.61

Previously

1.57

 sebelumnya

1.56

Activations Density 0.191%

phrases that indicate past events or actions

No Comments

No Known Activations