INDEX

Explanations

phrases that indicate a degree of moderation or slightness

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

wahlen

-0.62

 Fiske

-0.58

 nourrir

-0.54

portál

-0.54

ädie

-0.53

Carcinogenicity

-0.52

 Washburn

-0.52

ahalli

-0.51

 Harms

-0.51

 esclavos

-0.51

POSITIVE LOGITS

 bisschen

1.20

bit

1.09

 biraz

1.06

 المعيارى

1.03

 trochę

1.02

 biß

0.94

 کمی

0.93

 lidt

0.87

 Etwas

0.86

Bit

0.85

Activations Density 0.071%

phrases that indicate a degree of moderation or slightness

No Comments

No Known Activations

phrases that indicate a degree of moderation or slightness

No Comments

No Known Activations