INDEX

Explanations

conjunctions and other linking words that imply connection or relationships between subjects

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

list items followed by context

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

BIÉN

-0.49

itself

-0.48

CPtr

-0.47

 Academy

-0.47

 Itself

-0.46

scrapy

-0.45

kloped

-0.44

 itself

-0.43

 useStyles

-0.43

BIN

-0.43

POSITIVE LOGITS

 externi

0.57

tagHelperRunner

0.47

 يتيمه

0.46

NewUrlParser

0.45

 ProtoMessage

0.39

 numerus

0.39

Datuak

0.39

şört

0.38

 MULTIPLE

0.37

ABILIDAD

0.36

Activations Density 0.000%

conjunctions and other linking words that imply connection or relationships between subjects

list items followed by context

No Comments

No Known Activations