INDEX

Explanations

terms related to health metrics and chronic 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

 Luckily

-0.61

 silly

-0.59

Luckily

-0.57

beha

-0.57

ligent

-0.57

 luckily

-0.56

 Somehow

-0.55

OrDefault

-0.54

ligible

-0.53

 Silly

-0.53

POSITIVE LOGITS

expandindo

0.65

richTextPanel

0.63

исленность

0.60

abstractmethod

0.57

UnknownFields

0.56

 trebui

0.55

URLException

0.54

InjectAttribute

0.54

 îns

0.53

setDo

0.53

Activations Density 0.136%

terms related to health metrics and chronic conditions

No Comments

No Known Activations