INDEX

Explanations

phrases that indicate comparisons or differences

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

the only

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 each

-0.37

 surface

-0.34

 much

-0.34

 through

-0.33

-0.32

 broadly

-0.32

 these

-0.32

름

-0.31

 many

-0.31

데

-0.31

POSITIVE LOGITS

 tartalomajánló

0.85

 esternos

0.68

satunya

0.67

 nonUne

0.66

Only

0.64

 defaultstate

0.64

 Only

0.62

webElementXpaths

0.61

 Roskov

0.60

ONLY

0.60

Activations Density 0.026%

phrases that indicate comparisons or differences

the only

No Comments

No Known Activations