INDEX

Explanations

adjectives indicating high quality or degree

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 Engel

-0.71

 Beards

-0.69

jercicios

-0.65

 absc

-0.65

 térmica

-0.64

shah

-0.64

 scrubs

-0.63

 Gelb

-0.61

 Salter

-0.60

 hoạch

-0.60

POSITIVE LOGITS

 highly

2.52

highly

2.44

Highly

2.34

 Highly

2.34

 altamente

1.72

 extremely

1.25

extremely

1.15

Extremely

1.12

 hugely

1.11

 Extremely

1.07

Activations Density 0.049%

adjectives indicating high quality or degree

No Comments

No Known Activations