INDEX

Explanations

elements that indicate various forms of inquiry or reference in the context of processes, trends, and comparisons

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

referring to specific situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

AutoScale

-0.54

InjectMocks

-0.53

 betweenstory

-0.52

 lateinit

-0.46

 esterna

-0.46

Brainz

-0.45

énario

-0.45

 للاسماء

-0.45

このような

-0.43

 jScrollPane

-0.43

POSITIVE LOGITS

That

0.59

 That

0.57

 THAT

0.55

 بيها

0.52

在那

0.50

 thats

0.50

THAT

0.50

 Thats

0.48

 gyhoeddwyd

0.47

 staffers

0.46

Activations Density 0.143%

elements that indicate various forms of inquiry or reference in the context of processes, trends, and comparisons

referring to specific situations

No Comments

No Known Activations