INDEX

Explanations

specific terms related to technical and scientific contexts

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

for example

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

 Paglinawan

-0.73

endphp

-0.56

ſelf

-0.55

 itſelf

-0.54

 HasFactory

-0.54

abestanden

-0.54

twimg

-0.52

 Picchu

-0.50

 poffe

-0.50

DoubleQuotes

-0.50

POSITIVE LOGITS

 such

0.51

比如

0.49

 såsom

0.48

例如

0.48

StoryboardSegue

0.46

such

0.44

如

0.44

如

0.41

譬

0.40

 telles

0.39

Activations Density 0.107%

specific terms related to technical and scientific contexts

for example

No Comments

No Known Activations