INDEX

Explanations

instances of the word "wrote" and its variations, indicating discussions about writing or authorship

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

awtextra

-0.66

 légales

-0.60

PROLOG

-0.59

Bef

-0.57

 ujednoznacz

-0.56

 Quelles

-0.55

bei

-0.55

 Comparison

-0.54

Див

-0.53

 gắng

-0.51

POSITIVE LOGITS

 wrote

2.66

 writing

2.48

 Writing

2.20

writing

2.13

wrote

2.12

Writing

2.09

 WRITING

2.06

 Wrote

1.99

 written

1.92

 writings

1.89

Activations Density 0.136%

instances of the word "wrote" and its variations, indicating discussions about writing or authorship

No Comments

No Known Activations