INDEX

Explanations

expressions that indicate negation or limitation in argumentation

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

"let" followed by "alone"

np_acts-logits-general · gemini-2.0-flash

let alone, much less

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MLLoader

-0.68

soort

-0.66

TypedDataSet

-0.63

iastes

-0.60

Autoritní

-0.59

 FANDOM

-0.58

}();

-0.57

parsedMessage

-0.57

 Ammon

-0.57

 Wicidata

-0.57

POSITIVE LOGITS

let

2.11

let

1.61

 much

1.55

much

1.43

Let

1.40

Let

1.38

LET

1.27

 Much

1.24

Much

1.24

LET

1.23

Activations Density 0.181%

expressions that indicate negation or limitation in argumentation

"let" followed by "alone"

let alone, much less

No Comments

No Known Activations