INDEX

Explanations

references to programming and code-related operations

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

Text preceding common conjunctions

np_acts-logits-general · gemini-2.0-flash

conjunctions and subsequent phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Савезне

-0.86

fromnode

-0.81

 تضيفلها

-0.76

tagHelperRunner

-0.68

 kaarangay

-0.65

そのため

-0.64

 samot

-0.59

 TAMBÉM

-0.59

何より

-0.58

Билгалдахарш

-0.58

POSITIVE LOGITS

it

0.93

 they

0.78

发现

0.66

發現

0.58

It

0.57

 still

0.57

 оно

0.55

իմ

0.55

 everything

0.54

 found

0.54

Activations Density 0.216%

references to programming and code-related operations

Text preceding common conjunctions

conjunctions and subsequent phrases

No Comments

No Known Activations