INDEX

Explanations

technical or procedural language related to systems and methods

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

preference and priority

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

RenderAtEndOf

-0.54

ніципалі

-0.52

@"/

-0.45

intios

-0.38



-0.38

exitRule

-0.36

__;

-0.35

clairage

-0.35

 externes

-0.35

ciatore

-0.35

POSITIVE LOGITS

 preference

2.64

 Preference

2.33

 prefer

2.30

preference

2.22

prefer

2.19

 Prefer

2.17

 preferring

2.14

 prefers

2.06

 preferred

2.05

Preference

2.03

Activations Density 0.466%

technical or procedural language related to systems and methods

preference and priority

No Comments

No Known Activations