INDEX

Explanations

phrases that relate to responses and connections in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pleaſure

-0.98

 Theſe

-0.98

 Monfieur

-0.97

 Diſ

-0.97

 itſelf

-0.94

 Jefus

-0.93

 Chriftian

-0.92

 houſe

-0.92

Efq

-0.90

 purpoſe

-0.88

POSITIVE LOGITS

due

0.86

 because

0.84

 pursuant

0.81

μφωνα

0.76

 debido

0.74

 according

0.73

 вслед

0.72

 owing

0.68

dientemente

0.66

because

0.65

Activations Density 0.450%

phrases that relate to responses and connections in context

No Comments

No Known Activations