INDEX

Explanations

spatial relationships and orientations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

帖最后由

-0.65

IBAction

-0.57

@[+][

-0.55

unknownFields

-0.50

 Bonn

-0.50

MouseMove

-0.50

 Durant

-0.50

canestro

-0.49

 mutton

-0.49

 Swat

-0.47

POSITIVE LOGITS

 upside

1.02

 flipped

0.93

 Upside

0.90

 inverted

0.80

 flip

0.71

FLIP

0.71

 flips

0.71

 inversion

0.71

Flip

0.70

 Flip

0.70

Activations Density 0.034%

spatial relationships and orientations

No Comments

No Known Activations