INDEX

Explanations

verbs indicating actions related to investigations or analyses in scientific contexts

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

 happening

-0.68

happ

-0.59

nmgp

-0.58

DialogContent

-0.57

 acontecer

-0.55

DefaultValue

-0.52

istic

-0.52

 acontec

-0.51

ようになる

-0.49

 يتيمه

-0.49

POSITIVE LOGITS

LookAnd

0.84

fjspx

0.77

 subject

0.74

 Roskov

0.71

AddHtmlAttribute

0.70

 characterised

0.68

 characterized

0.67

subject

0.66

 NUKAT

0.65

 analysed

0.62

Activations Density 1.187%

verbs indicating actions related to investigations or analyses in scientific contexts

No Comments

No Known Activations