INDEX

Explanations

references to circuits 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 @ 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

 Danilo

-0.91

 kary

-0.90

 amitié

-0.88

+#+

-0.88

 raiſ

-0.88

Попис

-0.86

 crdi

-0.84

ytale

-0.83

ьаж

-0.83

 médicaux

-0.82

POSITIVE LOGITS

 circuit

2.25

 circuits

2.14

 Circuit

2.02

circuit

1.98

 Circuits

1.93

Circuit

1.85

 CIRCUIT

1.83

circuits

1.80

 circuito

1.55

 circuitos

1.47

Activations Density 0.127%

references to circuits in various contexts

No Comments

No Known Activations