INDEX

Explanations

phrases indicating propositions or descriptions within a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 myſelf

-1.23

 itſelf

-1.22

ніципалі

-1.20

 Monfieur

-1.20

Efq

-1.15

 himſelf

-1.10

 uſed

-1.09

 poffible

-1.08

 leſs

-1.07

 Chriftian

-1.05

POSITIVE LOGITS

<eos>

0.52

ś

0.40

Eff

0.40

↵↵

0.37

WindowEvent

0.35

seg

0.35

bra

0.35

Court

0.35

0.34

or

0.34

Activations Density 0.104%

phrases indicating propositions or descriptions within a technical context

No Comments

No Known Activations