INDEX

Explanations

coding functions and method invocations related to setting values or properties

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

 itſelf

-0.76

 Jefus

-0.70

сылкі

-0.69

Mediabestanden

-0.69

ſelf

-0.68

 poffible

-0.68

 Chriftian

-0.68

 contextLoads

-0.67

ſelves

-0.67

 pleaſure

-0.67

POSITIVE LOGITS

stdc

0.62

Hozzáférés

0.54

regate

0.53

Santis

0.53

Defaults

0.52

emp

0.52

Sol

0.52

στή

0.51

Sol

0.51

tting

0.50

Activations Density 0.071%

coding functions and method invocations related to setting values or properties

No Comments

No Known Activations