INDEX

Explanations

negations and their variations

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

enumi

-0.66

unicaciones

-0.56

 chaves

-0.52

 later

-0.50

PathVariable

-0.49

tonode

-0.49

çim

-0.48

 almofada

-0.48

 only

-0.47

IANGLES

-0.47

POSITIVE LOGITS

 istore

0.53

expandindo

0.46

WithIOException

0.46

jerry

0.45

 snippetHide

0.44

emper

0.44

fastjson

0.43

 LinearLayout

0.42

setupUi

0.42

[{\

0.41

Activations Density 0.017%

negations and their variations

No Comments

No Known Activations