INDEX

Explanations

actions that involve stepping or moving forward

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

 defStyleAttr

-0.53

tvrt

-0.51

cienti

-0.51

 Ärz

-0.50

 Palae

-0.48

intersects

-0.48

Abp

-0.48

 Fichier

-0.48

 Numerade

-0.47

 poffe

-0.47

POSITIVE LOGITS

 Stepping

1.55

 stepping

1.52

Stepping

1.49

 step

1.48

 stepped

1.43

 steps

1.41

 Step

1.39

 Steps

1.36

stepping

1.34

step

1.33

Activations Density 0.058%

actions that involve stepping or moving forward

No Comments

No Known Activations