INDEX

Explanations

terms related to legal or formal disclosures

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

relative pronouns

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

また

-0.35

 Autorisations

-0.34

anstalt

-0.33

发表于

-0.32

kool

-0.31

kolen

-0.30

enfone

-0.29

ifikationen

-0.29

äler

-0.29

 tắt

-0.29

POSITIVE LOGITS

 which

2.30

 whose

1.72

which

1.71

whose

1.56

 whom

1.45

 которых

1.45

 cuales

1.42

 которого

1.41

 cuyo

1.40

 WHICH

1.38

Activations Density 0.068%

terms related to legal or formal disclosures

relative pronouns

No Comments

No Known Activations