INDEX

Explanations

repetitive uses of the word "had" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

LLocation

-0.96

 houſe

-0.93

 perfons

-0.87

 perſon

-0.87

 ſtate

-0.86

 Erscheinung

-0.86

PreferredItem

-0.86

 occafion

-0.85

 Torrey

-0.85

 Ressources

-0.85

POSITIVE LOGITS

had

1.32

Had

1.28

HAD

1.23

Had

1.14

had

1.00

HAD

0.86

0.84

Earlier

0.83

 earlier

0.80

 hoped

0.80

Activations Density 0.095%

repetitive uses of the word "had" in various contexts

No Comments

No Known Activations