INDEX

Explanations

expressions of opinions and viewpoints

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 illustrazione

-0.65

 forbindelse

-0.53

AddTagHelper

-0.50

 verwijzen

-0.49

上角

-0.49

dstuk

-0.48

 cipolla

-0.47

זכ

-0.46

 кӀош

-0.45

 proprement

-0.44

POSITIVE LOGITS

 opinions

1.67

 opinion

1.66

 Opinions

1.46

opinion

1.39

Opinions

1.39

opinions

1.36

 Opinion

1.36

Opinion

1.27

 OPINION

1.18

意見

1.15

Activations Density 0.728%

expressions of opinions and viewpoints

No Comments

No Known Activations