INDEX

Explanations

instances of the word "Har" and its variations

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

XmlType

-0.58

 дописавши

-0.58

 sumpay

-0.57

 mergeFrom

-0.56

 Lenz

-0.56

onAttach

-0.55

EDEFAULT

-0.55

 članak

-0.54

помним

-0.53

ForValue

-0.52

POSITIVE LOGITS

HAR

0.82

 cardiaque

0.81

Har

0.79

har

0.79

Har

0.72

 Harlow

0.70

har

0.67

HAR

0.67

 harvester

0.65

 Harrell

0.65

Activations Density 0.079%

instances of the word "Har" and its variations

No Comments

No Known Activations