INDEX

Explanations

phrases indicating positional or situational contexts

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

phrases indicating position or readiness for action

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

in charge of

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

abstractmethod

-0.58

aarrggbb

-0.58

ostante

-0.57

GeneratedValue

-0.57

TypedDataSet

-0.56

 ['./

-0.56

 архивлан

-0.54

AutoScale

-0.54

améli

-0.52

 nakalista

-0.50

POSITIVE LOGITS

SourceChecksum

0.59

órum

0.53

 versteht

0.53

__*/

0.53

jects

0.53

RSpec

0.52

 sometimes

0.51

umeric

0.50

 everything

0.50

 strong

0.50

Activations Density 0.064%

phrases indicating positional or situational contexts

phrases indicating position or readiness for action

in charge of

No Comments

No Known Activations