INDEX

Explanations

attributes related to analysis and identification processes

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

ValueStyle

-0.65

-0.62

Jegyzetek

-0.59

IsContent

-0.56

oa̍t

-0.54

 صوتيه

-0.53

Aholisi

-0.53

 vectorielles

-0.53

hasMoreElements

-0.52

orative

-0.52

POSITIVE LOGITS

 anteriormente

0.65

 heretofore

0.64

 حديث

0.64

 therein

0.62

 earlier

0.61

 appunto

0.61

 herein

0.61

WebServlet

0.60

 herewith

0.60

 Earlier

0.60

Activations Density 0.669%

attributes related to analysis and identification processes

No Comments

No Known Activations