INDEX

Explanations

phrases indicating confirmation or support of previous research or findings

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

 Diſ

-0.81

 ―――――

-0.81

 greateſt

-0.81

 Majefty

-0.81

 Perſ

-0.79

ſelf

-0.76

ſelves

-0.74

 preſent

-0.74

 Conſ

-0.73

 Jefus

-0.72

POSITIVE LOGITS

 previous

0.92

previous

0.82

 предыду

0.75

earlier

0.75

 précédente

0.74

 anteriores

0.71

 précédentes

0.68

 earlier

0.68

 recent

0.68

Previous

0.64

Activations Density 0.851%

phrases indicating confirmation or support of previous research or findings

No Comments

No Known Activations