INDEX

Explanations

programming and coding syntax

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

цах

-0.41

тере

-0.41

rih

-0.39

<eos>

-0.39

uten

-0.38

Had

-0.38

ble

-0.37

 made

-0.36

час

-0.36

POSITIVE LOGITS

rungsseite

0.98

 propOrder

0.87

]--;

0.87

findpost

0.80

medriver

0.75

niająca

0.75

 Wicidata

0.74

NUMX

0.73

Efq

0.73

MLLoader

0.72

Activations Density 0.410%

programming and coding syntax

No Comments

No Known Activations