INDEX

Explanations

concepts related to measurement or comparison of conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

MemoryWarning

-0.65

 iprot

-0.63

 CURIAM

-0.62

elemField

-0.59

 propOrder

-0.54

complexContent

-0.53

utuhkan

-0.53

+#+

-0.52

INSEE

-0.52

tableFuture

-0.51

POSITIVE LOGITS

 remain

0.84

 stays

0.77

 remains

0.76

 remained

0.71

 fica

0.70

 becomes

0.68

 fique

0.65

 become

0.64

 Remain

0.64

 Remains

0.62

Activations Density 0.757%

concepts related to measurement or comparison of conditions

No Comments

No Known Activations