INDEX

Explanations

phrases that express causal relationships or consequences in arguments

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

 متعلقه

-0.85

 Grâce

-0.83

GEBURTSDATUM

-0.82

yntaxException

-0.82

الحياه

-0.81

 للاسماء

-0.79

 propOrder

-0.79

 autorytatywna

-0.78

featureID

-0.77

 صوتيه

-0.72

POSITIVE LOGITS

Certainly

1.04

 Certainly

1.00

 certainly

0.90

certainly

0.86

 certamente

0.80

 Indeed

0.80

 even

0.80

Indeed

0.76

 Presumably

0.74

 indeed

0.71

Activations Density 1.271%

phrases that express causal relationships or consequences in arguments

No Comments

No Known Activations