INDEX

Explanations

instances of the word "voice" and its variations to identify discussions centered around personal expression and agency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

glers

-0.79

Tembelea

-0.76

weds

-0.75

μβρίου

-0.74

CrossRef

-0.73

]").

-0.71

edicated

-0.71

 monasteries

-0.69

 communs

-0.68

✷

-0.67

POSITIVE LOGITS

 voice

2.23

 voices

1.92

 Voice

1.92

voice

1.86

Voice

1.83

 VOICE

1.82

 Voices

1.66

VOICE

1.63

voices

1.58

Voices

1.52

Activations Density 0.058%

instances of the word "voice" and its variations to identify discussions centered around personal expression and agency

No Comments

No Known Activations