INDEX

Explanations

phrases related to requesting additional information

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

Followed by actions like "visit," "go," "read," or "call."

np_acts-logits-general · gemini-2.0-flash

information followed by action

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

Hopefully

-0.61

RepeatedField

-0.60

 zulassen

-0.58

contentLoaded

-0.57

WithIOException

-0.56

 roulant

-0.56

 يتيمه

-0.55

 Hopefully

-0.54

RemindMe

-0.53

ďaka

-0.52

POSITIVE LOGITS

 visit

1.71

visit

1.31

 Visit

1.18

 VISIT

1.17

Visit

1.13

 visita

1.07

 visited

1.07

go

1.05

 visits

1.02

 visite

1.02

Activations Density 0.119%

phrases related to requesting additional information

Followed by actions like "visit," "go," "read," or "call."

information followed by action

No Comments

No Known Activations