INDEX

Explanations

features related to product specifications and usability

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

endencies

-0.56

 Systems

-0.55

 poveznice

-0.53

mybatisplus

-0.53

BufferException

-0.51

 Jüng

-0.51

Dislikes

-0.51

RenderAtEndOf

-0.49

toHaveBeenCalled

-0.49

😦

-0.49

POSITIVE LOGITS

 lightweight

0.69

 durable

0.68

 sturdy

0.60

lightweight

0.59

 compact

0.59

 chi̍t

0.58

 high

0.58

 simple

0.57

durable

0.56

Lightweight

0.56

Activations Density 0.039%

features related to product specifications and usability

No Comments

No Known Activations