INDEX

Explanations

references to people, particularly in terms of personal pronouns and their significance 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 @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 cauſe

-1.14

 pleaſure

-1.07

 purpoſe

-1.06

 uſe

-0.99

 houſe

-0.95

 fevere

-0.95

 Houſe

-0.95

 raiſ

-0.94

 reaſon

-0.94

 caufe

-0.93

POSITIVE LOGITS

self

0.87

us

0.83

HIM

0.74

us

0.73

him

0.73

Them

0.71

 them

0.69

 Them

0.69

Him

0.68

me

0.68

Activations Density 0.139%

references to people, particularly in terms of personal pronouns and their significance in context

No Comments

No Known Activations