INDEX

Explanations

relationships and their implications in various contexts

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

introducing lists of items

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

 AssemblyCompany

-0.43

 natomiast

-0.40

Reprodução

-0.36

却

-0.35

verket

-0.33

SpringBootTest

-0.33

ür

-0.32

脱

-0.32

XmlSchema

-0.31

 steig

-0.31

POSITIVE LOGITS

SequentialGroup

0.61

mybatisplus

0.60

Including

0.57

INCLUDING

0.53

 incluindo

0.52

 nakalista

0.52

Билгалдахарш

0.52

الحياه

0.52

including

0.51

новништво

0.51

Activations Density 0.052%

relationships and their implications in various contexts

introducing lists of items

No Comments

No Known Activations