INDEX

Explanations

temporal transitions and indicators in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 EconPapers

-0.56

 ſtate

-0.55

 Grecs

-0.50

 houſe

-0.49

 jScrollPane

-0.48

ulemon

-0.48

 ſche

-0.48

Pozri

-0.48

 Atlántico

-0.47

AppMethodBeat

-0.47

POSITIVE LOGITS

 afterward

0.81

 danach

0.80

 afterwards

0.80

 subsequently

0.78

Затем

0.77

 anschließend

0.75

 násled

0.75

 thereafter

0.73

 Afterwards

0.71

 Затем

0.71

Activations Density 0.440%

temporal transitions and indicators in text

No Comments

No Known Activations