INDEX

Explanations

punctuations and specific conversational markers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

今後も

-0.69

これからも

-0.68

wnież

-0.62

Références

-0.62

Demikian

-0.61

 Jeśli

-0.60

قایناقلار

-0.59

 Semoga

-0.59

częściej

-0.59

 disponibilités

-0.58

POSITIVE LOGITS

Turns

1.20

 Turns

1.14

 turns

0.99

turns

0.84

 guess

0.82

 immediately

0.81

 Immediately

0.78

Immediately

0.75

 Turned

0.75

 Upon

0.74

Activations Density 0.234%

punctuations and specific conversational markers

No Comments

No Known Activations