INDEX

Explanations

references to candidates and their qualifications in various contexts

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

candidate's identity or position

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

хьтан

-0.60

hoeddwyd

-0.57

 виправивши

-0.55

 surla

-0.54

Jîn

-0.54

HomeAsUp

-0.53

 NSCoder

-0.52

蚪

-0.50

usermodel

-0.49

WithIOException

-0.48

POSITIVE LOGITS

 sepen

0.40

isContained

0.40

ſelves

0.38

 évêque

0.38

 Conſ

0.36

BufferException

0.36

izarse

0.36

 doğum

0.36

 Reſ

0.36

ſelf

0.35

Activations Density 0.151%

references to candidates and their qualifications in various contexts

candidate's identity or position

No Comments

No Known Activations