INDEX

Explanations

terms indicating clear definitions and specifications

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

definite, defined, specific, concrete

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

TagMode

-0.60

存于互联网档案馆

-0.52

 Panto

-0.45

ьаж

-0.41

 Drag

-0.41

 NSObject

-0.41

 beginnetje

-0.41

multirow

-0.40

 Highlight

-0.40

 true

-0.40

POSITIVE LOGITS

StreetMap

0.43

 nonUne

0.41

 sólida

0.41

 padat

0.39

 sólido

0.39

 konkre

0.39

endcsname

0.38

 specificity

0.37

แน่น

0.37

setVerticalGroup

0.37

Activations Density 0.106%

terms indicating clear definitions and specifications

definite, defined, specific, concrete

No Comments

No Known Activations