INDEX

Explanations

phrases indicating intentions or actions aimed at achieving specific goals

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ſtate

-0.89

 ftate

-0.89

 Majefty

-0.83

Efq

-0.80

 fubject

-0.79

 houſe

-0.79

 pleaſure

-0.78

 whoſe

-0.77

 chofe

-0.76

 poffe

-0.73

POSITIVE LOGITS

 Để

0.93

 чтобы

0.88

为了

0.86

 כדי

0.85

 afin

0.85

Cyfeiriadau

0.82

Để

0.82

ůli

0.81

 Чтобы

0.80

為了

0.79

Activations Density 0.096%

phrases indicating intentions or actions aimed at achieving specific goals

No Comments

No Known Activations