INDEX

Explanations

references to intellectual discourse and cultural analysis

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

like "likewise", "similarly", or "equivalent"

np_acts-logits-general · gemini-2.0-flash

likewise, equivalent, similar

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 kasarigan

-0.65

ReusableCell

-0.63

 transfieras

-0.61

VYMaps

-0.61

verifyException

-0.60

AsUp

-0.59

oa̍t

-0.59



-0.58

balleur

-0.58

InitVars

-0.57

POSITIVE LOGITS

 podob

0.41

像

0.36

plaat

0.35

Operation

0.35

 equivalent

0.34

 similares

0.34

 equivalents

0.34

 equally

0.33

 analogous

0.33

 Likewise

0.32

Activations Density 1.273%

references to intellectual discourse and cultural analysis

like "likewise", "similarly", or "equivalent"

likewise, equivalent, similar

No Comments

No Known Activations