INDEX

Explanations

references to specific entities and relationships in a political or legal context

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

observations and code

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 betweenstory

-0.33

ToBe

-0.29

 dié

-0.29

 Throwable

-0.28

 initView

-0.28

FILED

-0.28

 Consultancy

-0.27

 callable

-0.27

leetcode

-0.27

POSITIVE LOGITS

AndEndTag

0.90

 BoxFit

0.68

ValueStyle

0.65

ſammen

0.60

ExtendWith

0.60

 sumpay

0.60

WebVitals

0.59

 createSlice

0.56

ftagPool

0.55

 utafitiHapana

0.55

Activations Density 0.004%

references to specific entities and relationships in a political or legal context

observations and code

No Comments

No Known Activations