INDEX

Explanations

references to the significance of various concepts or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

InjectAttribute

-0.76

Портал

-0.68

 Ginger

-0.52

Cham

-0.51

 przew

-0.51

بوابة

-0.50

Ainsi

-0.50

 Cuth

-0.49

mink

-0.49

piram

-0.49

POSITIVE LOGITS

={()

0.92

 importance

0.72

0.71

={()=>

0.70

praš

0.66

importance

0.65

őbb

0.63

Importance

0.63

pic

0.62

 mahar

0.61

Activations Density 0.085%

references to the significance of various concepts or issues

No Comments

No Known Activations