INDEX

Explanations

references to previously mentioned or described information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

исленность

-0.37

udder

-0.37

 cruiser

-0.37

otic

-0.37

 Jesu

-0.36

OnAdd

-0.36

erializer

-0.35

 fete

-0.35

 cruisers

-0.35

liev

-0.35

POSITIVE LOGITS

 mentioned

0.72

 above

0.71

 listed

0.69

 outlined

0.68

 shown

0.66

 described

0.63

above

0.61

shown

0.60

 diatas

0.59

 mencionados

0.59

Activations Density 1.105%

references to previously mentioned or described information

No Comments

No Known Activations