INDEX

Explanations

phrases indicating prior actions or events related to participation and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 متعلقه

-0.73

 actuelles

-0.66

cephala

-0.65

lehem

-0.62

attuale

-0.62

woordig

-0.62

étaire

-0.61

 actuelle

-0.60

perity

-0.59

ibouti

-0.58

POSITIVE LOGITS

 earlier

1.12

 previously

1.03

 hadn

0.93

Earlier

0.92

earlier

0.91

 Earlier

0.89

 previous

0.86

had

0.86

 zuvor

0.85

 anteriormente

0.83

Activations Density 0.459%

phrases indicating prior actions or events related to participation and events

No Comments

No Known Activations