INDEX

Explanations

references to data structures and numerical parameters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ệc

-0.63

 pleaſure

-0.61

новништво

-0.60

 насељу

-0.60

下载附件

-0.59

 preuves

-0.57

interopRequire

-0.56

Espèce

-0.55

 nhat

-0.55

FromJson

-0.54

POSITIVE LOGITS

0.71

0.65

0.63

0.60

0.58

0.55

 Career

0.54

0.51

hijab

0.50

memoized

0.50

Activations Density 0.337%

references to data structures and numerical parameters

No Comments

No Known Activations