INDEX

Explanations

expressions of personal reflection and existential questioning

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

following "while," "was," or "were"

np_acts-logits-general · gemini-2.0-flash

while/was/I followed by physical 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

PasswordEncoder

-0.76

 EconPapers

-0.69

 SwitchCompat

-0.61

ďaka

-0.58

>?

-0.52

Enllaces

-0.51

optString

-0.50

"]];

-0.49

getOriginal

-0.49

 verkar

-0.48

POSITIVE LOGITS

 sitting

1.73

 Sitting

1.47

Sitting

1.46

sitting

1.45

 standing

1.36

 walking

1.33

 seated

1.22

sit

1.21

sat

1.17

 sentado

1.16

Activations Density 0.388%

expressions of personal reflection and existential questioning

following "while," "was," or "were"

while/was/I followed by physical action

No Comments

No Known Activations