INDEX

Explanations

verbs indicating actions or processes, especially in a scientific or procedural context

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

user prompt

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

 initComponents

-0.50

AddTagHelper

-0.43

complexType

-0.42

afficheront

-0.41

 Chwiliwch

-0.38

lankton

-0.36

 Exactos

-0.36

فایل‌لار

-0.35

 himo

-0.35

BufferException

-0.35

POSITIVE LOGITS

 مشين

0.59

Viitteet

0.49

 illet

0.48

 belonged

0.48

rungsseite

0.47

 coordonnées

0.46

 esmal

0.45

 mijne

0.45

 téléphone

0.45

XMLSchema

0.45

Activations Density 0.000%

verbs indicating actions or processes, especially in a scientific or procedural context

user prompt

No Comments

No Known Activations