INDEX

Explanations

repetitive phrases related to existence or state in various contexts

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

|()

-0.59

illary

-0.52

Conversely

-0.51

 occa

-0.51

amqp

-0.51

 glean

-0.47

lettes

-0.47

 ostensibly

-0.46

 Conversely

-0.46

assioned

-0.46

POSITIVE LOGITS

 كومونز

0.79

SourceChecksum

0.72

 very

0.70

 advices

0.69

 gonna

0.68

 prouve

0.68

 rempla

0.67

 intptr

0.63

 pasando

0.63

Ikr

0.62

Activations Density 0.798%

repetitive phrases related to existence or state in various contexts

No Comments

No Known Activations