INDEX

Explanations

names or references to specific individuals or characters

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

GEBURTSDATUM

-0.71

djangoproject

-0.64

awaiter

-0.63

adaptiveStyles

-0.61

invokeLater

-0.61

Personensuche

-0.59

帖最后由

-0.59

FDRE

-0.58

⎢

-0.57

webElementXpaths

-0.57

POSITIVE LOGITS

hhhh

0.64

 Luch

0.63

Lach

0.63

hhhhhhhh

0.61

 Lach

0.57

HHHH

0.56

lorette

0.55

ado

0.55

HHHHHHHH

0.55

hhhhh

0.55

Activations Density 0.213%

names or references to specific individuals or characters

No Comments

No Known Activations