INDEX

Explanations

references to historical events and periods

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

`respectively` `too` `later` `subsequent`

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

 potential

-0.64

styleType

-0.62

 Cæsar

-0.61

 kaarangay

-0.60

 skills

-0.59

Identyfik

-0.59

 advice

-0.58

 needs

-0.58

 satisfaction

-0.56

NUMX

-0.55

POSITIVE LOGITS

 furent

0.73

 assassinated

0.65

 posteriores

0.65

 infamous

0.63

 Historical

0.62

 posteriormente

0.61

 Later

0.61

 Subsequent

0.59

Later

0.59

 '\\;'

0.57

Activations Density 0.063%

references to historical events and periods

`respectively` `too` `later` `subsequent`

No Comments

No Known Activations