INDEX

Explanations

details related to personal names and identifying information

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

Appearing before names or job titles

np_acts-logits-general · gemini-2.0-flash

titles and names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 ComVisible

-0.74

FieldBuilder

-0.66

 ThemeData

-0.63

 typelib

-0.63

AndEndTag

-0.63

SequentialGroup

-0.57

NgModule

-0.55

ContentAsync

-0.55

setVerticalGroup

-0.55

principalTable

-0.54

POSITIVE LOGITS

PhD

0.93

 Senior

0.88

 senior

0.83

Jr

0.82

 Assistant

0.80

Senior

0.79

Esq

0.79

 Associate

0.76

senior

0.76

 founder

0.75

Activations Density 0.369%

details related to personal names and identifying information

Appearing before names or job titles

titles and names

No Comments

No Known Activations