INDEX

Explanations

references to conceptual or abstract ideas about 'things'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 VIAF

-0.60

letas

-0.47

me

-0.46

ngrx

-0.45

 onbe

-0.43

 avantages

-0.43

sue

-0.43

(({

-0.43

wiad

-0.42

Iden

-0.42

POSITIVE LOGITS

Things

0.96

 things

0.93

 Things

0.90

things

0.86

 THINGS

0.81

 disambiguazione

0.78

ArrowToggle

0.77

 createSlice

0.75

ItemBackground

0.74

THINGS

0.71

Activations Density 0.138%

references to conceptual or abstract ideas about 'things'

No Comments

No Known Activations