INDEX

Explanations

concepts related to causal and analytical frameworks in research

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

preceding "to be" or "in question"

np_acts-logits-general · gemini-2.0-flash

being involved

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

abetes

-0.71

StoryboardSegue

-0.65

исленность

-0.63

 layui

-0.59

ChildScrollView

-0.59

EDEFAULT

-0.57

перь

-0.56

TRAILING

-0.56

complexContent

-0.56

 Normdatei

-0.56

POSITIVE LOGITS

 involved

1.50

 being

1.36

 concerned

1.27

involved

1.17

 Involved

1.17

concerned

1.13

being

1.09

 INVOLVED

1.02

Being

1.01

 BEING

1.00

Activations Density 0.932%

concepts related to causal and analytical frameworks in research

preceding "to be" or "in question"

being involved

No Comments

No Known Activations