INDEX

Explanations

references or citations in the text

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

referring to previous statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

eload

-0.50

odha

-0.49

jspb

-0.46

partements

-0.45

 Moral

-0.45

existsSync

-0.44

Kne

-0.43

 ventilation

-0.43

 temptation

-0.43

 Clerical

-0.42

POSITIVE LOGITS

 nakalista

0.52

Hochspringen

0.51

 ***!

0.48

 atlik

0.46

 gyhoeddwyd

0.45

TagMode

0.43

 незавершена

0.42

FormTagHelper

0.41

 auroit

0.41

上面

0.40

Activations Density 0.017%

references or citations in the text

referring to previous statements

No Comments

No Known Activations