INDEX

Explanations

temporal references in events or discussions

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

days and months

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 EconPapers

-0.68

 noDo

-0.66

 ſeine

-0.61

 CanadaChoose

-0.60

 ſei

-0.59

 queſta

-0.59

xase

-0.59

 ब्रेकडाउन

-0.59

 laſſen

-0.59

 verſch

-0.57

POSITIVE LOGITS

 Thursday

0.42

 Tuesday

0.42

 Wednesday

0.38

помним

0.35

 Monday

0.33

queryset

0.33

 Friday

0.32

Tuesday

0.32

Thursday

0.31

 today

0.31

Activations Density 0.009%

temporal references in events or discussions

days and months

No Comments

No Known Activations