INDEX

Explanations

concepts related to overlooked details and considerations in various contexts

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

following "been", "often", or "was" indicating something omitted

np_acts-logits-general · gemini-2.0-flash

overlooked or missed

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

DebuggerNonUser

-0.64

énario

-0.57

Chham

-0.52

PyExc

-0.51

 Merah

-0.49

 Genu

-0.48

 parseFrom

-0.48

日閲覧

-0.47

Portail

-0.46

GoogleApiClient

-0.46

POSITIVE LOGITS

 overlooked

2.29

 ignored

2.19

 neglected

1.99

 ignore

1.96

 overlook

1.93

 forgotten

1.92

 neglect

1.89

 ignoring

1.83

 ignores

1.76

 Ignore

1.73

Activations Density 0.623%

concepts related to overlooked details and considerations in various contexts

following "been", "often", or "was" indicating something omitted

overlooked or missed

No Comments

No Known Activations