INDEX

Explanations

phrases indicating intention or desire

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

VIAF

-0.39

NameInMap

-0.38

tangentMode

-0.37

 Permalink

-0.36

PropertyChanging

-0.36

CONCLUSIONES

-0.36

CONCLUSIONS

-0.35

 conclu

-0.35

+#+

-0.34

POSITIVE LOGITS

 whoſe

0.57

 pleaſure

0.56

ſelves

0.56

 ſta

0.54

 Theſe

0.53

ſelf

0.53

évaluateur

0.51

 leaſt

0.50

 Anſ

0.50

AddTagHelper

0.50

Activations Density 0.126%

phrases indicating intention or desire

No Comments

No Known Activations