INDEX

Explanations

terms related to admissions processes and criteria

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

admission to

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

 kasarigan

-0.70

carolina

-0.58

yarnpkg

-0.55

stdc

-0.55

errorCode

-0.55

Mek

-0.55

exca

-0.54

Pyrene

-0.54

schrank

-0.54

keletal

-0.54

POSITIVE LOGITS

 admission

0.97

 Admission

0.96

Admission

0.84

admission

0.75

 Admissions

0.69

 Admit

0.68

 admit

0.67

 admitted

0.67

 admissions

0.66

 admittance

0.64

Activations Density 0.008%

terms related to admissions processes and criteria

admission to

No Comments

No Known Activations