INDEX

Explanations

dialogues and quoted speech in the text

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

opening quote followed by pronoun

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

icity

-0.48

gridx

-0.44

ResumeLayout

-0.44

 Gele

-0.44

 tempting

-0.43

GenerationType

-0.43

bergen

-0.43

EndContext

-0.43

 consuma

-0.42

 hijack

-0.42

POSITIVE LOGITS

 AssemblyCompany

0.46

 transfieras

0.44

 invokingState

0.43

 athletic

0.40

 étoit

0.40

CrossRef

0.38

 propOrder

0.38

 선수

0.37

 psicológico

0.36

 hObject

0.36

Activations Density 0.004%

dialogues and quoted speech in the text

opening quote followed by pronoun

No Comments

No Known Activations