INDEX

Explanations

references to authors and researchers in academic writing

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 saisi

-0.58

tilizers

-0.56

日閲覧

-0.54

 Prompt

-0.54

 taches

-0.54

⬛

-0.54

matorium

-0.54

 cellspacing

-0.53

нивер

-0.52

τυ

-0.52

POSITIVE LOGITS

HHS

0.79

SequentialGroup

0.77

HRS

0.75

HTS

0.73

 kaarangay

0.73

HN

0.70

encodeWith

0.68

 Heroes

0.67

HX

0.67

 Hawks

0.66

Activations Density 1.596%

references to authors and researchers in academic writing

No Comments

No Known Activations