INDEX

Explanations

references to future timeframes or events

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 utafitiHapana

-0.45

 fieltro

-0.38

 kasarigan

-0.36

AddressBook

-0.35

дя

-0.35

 felicitación

-0.35

省市镇

-0.35

 noticia

-0.35

ParallelGroup

-0.35

 exemplar

-0.34

POSITIVE LOGITS

 purpoſe

0.57

 pleaſure

0.57

ſelf

0.54

 Reſ

0.52

 myſelf

0.51

writeFieldEnd

0.51

IVEREF

0.50

featureID

0.49

 pinulongan

0.48

 ſche

0.48

Activations Density 0.020%

references to future timeframes or events

No Comments

No Known Activations