INDEX

Explanations

questions related to perceived difficulties or challenges

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

Following "how," "can," "could," "would" or "conduction"

np_acts-logits-general · gemini-2.0-flash

How can you

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 StatefulWidget

-0.74

ſelves

-0.68

 Monfieur

-0.68

 itſelf

-0.65

 Majefty

-0.64

 auffi

-0.64

ſelf

-0.63

aarrggbb

-0.63

 myſelf

-0.63

 Audiodateien

-0.62

POSITIVE LOGITS

 skjø

0.52

timewa

0.49

Resumo

0.47

 rów

0.46

hame

0.46

 supone

0.44

lapi

0.44

pew

0.43

olidation

0.43

unsplash

0.43

Activations Density 0.164%

questions related to perceived difficulties or challenges

Following "how," "can," "could," "would" or "conduction"

How can you

No Comments

No Known Activations