INDEX

Explanations

terms that signify progress or achievement

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

 незавершена

-0.48

omsday

-0.48

 retaliation

-0.47

nola

-0.47

ret

-0.45

framework

-0.44

 useDispatch

-0.43

jenner

-0.43

wra

-0.43

ikir

-0.42

POSITIVE LOGITS

 proving

1.46

 prove

1.43

 proved

1.40

 proves

1.36

 proven

1.31

 Proving

1.19

Prove

1.16

 Prove

1.14

 demostrar

1.14

 prouver

1.09

Activations Density 0.294%

terms that signify progress or achievement

No Comments

No Known Activations