INDEX

Explanations

cooking instructions and process-related terms

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

وعة

-0.63

quelize

-0.53

IALE

-0.49

 append

-0.48

 waschen

-0.48

 FormBuilder

-0.48

tencent

-0.48

://$

-0.47

puted

-0.47

ungle

-0.47

POSITIVE LOGITS

 brown

0.64

 AssemblyTitle

0.60

 browning

0.58

ArrowToggle

0.56

MethodManager

0.55

 browned

0.55

insee

0.55

brown

0.54

0.53

riwal

0.53

Activations Density 0.130%

cooking instructions and process-related terms

No Comments

No Known Activations